FOOD AND DRUG ADMINISTRATION

P R O C E E D I N G S

(9:04 a.m.)

CHAIRPERSON HENDRICKS: Good morning. I'd like to call this meeting of the National Mammography Quality Assurance Advisory Committee to order.

I also request that everyone in attendance at this meeting sign in on the attendance sheet that is available at the door.

DR. FINDER: Okay. The following announcement addresses conflict of interest issues associated with this meeting and is made a par of the record to preclude even the appearance of any impropriety.

To determine if any conflict existed, the agency reviewed the submitted agenda and all financial interests reported by the committee participants. The conflict of interest statutes prohibits special government employees from participating in matters that could affect their or their employer's financial interests.

However, the agency has determined that participation of certain members, the need for whose services outweighs the potential conflict of interest involved is in the best interest of the government.

Therefore, waivers permitting full participation in general matters that come before the committee have been granted for certain participants because of their financial involvement with the facilities that will be subject to FDA's regulations on mammography quality standards with accrediting, certifying or inspecting bodies, with manufacturers of mammography equipment, or with their professional affiliations since these organizations could be affected by the committee's deliberations.

These individuals are Ms. Diane Rinella, Ms. Jacquelin Holland, Ms. Debra Monticciolo, Mr. William Passetti, Dr. Mark Williams, and Ms. Jane Segelken.

Waivers are currently on file for Dr. Carolyn Hendricks, Dr. Scott Ferguson, Ms. Carol Mount, Ms. Alisa Gilbert, Dr. Miles Harrison, Ms. Linda Pura, and Ms. Melissa Martin.

Copies of the waivers may be obtained from the agency's Freedom of Information Office, Room 12A-15 of the Parklawn Building.

We would like to note for the record that if any discussion of states or certifying bodies was to take place in any meetings of the committee, it would be a general discussion only. No vote would be taken and no consensus sought.

In the interest of getting as many viewpoints as possible all SGEs, including state employees, would be allowed to participate in the general discussion so that all viewpoints could be heard.

In the event that the discussions involve any other matters not already on the agenda in which an FDA participant has a financial interest, the participant should excuse him or herself from such involvement and the exclusion will be noted for the record.

With respect to all other participants, we ask in the interest of fairness that all persons making statements or presentations disclose any current or previous financial involvement with accreditation bodies, states doing mammography inspections under contract to FDA, certifying bodies, mobile units, breast implant imaging, consumer complaints, and mammography equipment.

CHAIRPERSON HENDRICKS: I note for the record that the voting members present constitute a quorum as required by 21 CFR Part 14.

At this time we'd like to move to the introduction of the panel members. Beginning from the right side, I'd like to have each member make a brief introduction.

MS. PURA: Good morning. I'm Linda Pura. I am Clinical Coordinator from the Los Angeles County Regional Partnership for Cancer Detection; also am a Susan G. Coleman Breast Cancer Foundation volunteer.

MS. HOLLAND: Good morning. My name is Jacquelin Holland, and I'm Program Director of the Diversity Enhancement Program at the James Cancer Hospital and Soloff (phonetic) Research Institute in Columbus Ohio.

MS. GILBERT: Good morning. I'm Alisa Gilbert from the Office of Native Cancer Survivorship in Alaska, Anchorage, Alaska.

DR. WILLIAMS: And I'm Mark Williams from the University of Virginia, and I'm an Associate Professor of Radiology, Biomedical Engineering and Physics there.

MS. SEGELKEN: I'm Jane Segelken. I'm a breast cancer survivor, and I'm a volunteer with the Ithaca Breast Cancer Alliance.

DR. MONTICCIOLO: Good morning. I'm Debbie Monticciolo. I'm a Professor of Radiology and Section Chief of Breast Imaging at Texas A&M in Temple, Texas.

DR. FERGUSON: I'm Scott Ferguson. I'm a diagnostic radiologist from the State of Arkansas.

MS. RINELLA: Good morning. I'm Diane Rinella, mammography technologist and consultant.

DR. FINDER: Charles Finder. I'm a radiologist. I work for the Food and Drug Administration. I'm the Executive Secretary of this Committee.

CHAIRPERSON HENDRICKS: I'm Carolyn Hendricks. I'm a medical oncologist in private practice, and I focus on breast disease, and I'm chairing this committee.

MR. PASSETTI: Bill Passetti. I'm Director of Florida's Radiation Control Agency from Tallahassee, Florida.

MS. MOUNT: I'm Carol Mount, a manager of the Breast Imaging and Intervention Center, Mayo Clinic, Rochester, Minnesota.

DR. MARTIN: Melissa Martin. I'm a consulting medical physicist in Southern California area.

DR. FINDER: Dr. Harrison is coming in by telephone teleconference. Dr. Harrison?

DR. HARRISON: Yes. Good morning. I'm Miles Harrison of Baltimore, Maryland, a breast surgeon.

DR. FINDER: Thank you.

CHAIRPERSON HENDRICKS: At this time I'd like to make a brief statement specifically addressed to the individuals who will be speaking in the open public hearing sections of this meeting.

Both the FDA and the public believe in a transparent process for information gathering and decision making. To insure such transparency at the open public hearing session of this Advisory Committee, FDA believes it is important to understand the context of an individual's presentation. For this reason, FDA encourages you, the open public hearing speaking at the beginning of your written or oral statement to advise this committee of any financial relationship that you may have with the sponsor, its product, and if know, its direct competitors.

For example, this financial information may include the sponsor's payment of your travel, lodging or other expenses in connection with your attendance at this meeting.

Likewise, FDA encourages you at the beginning of your statement to advise this committee if you do not have any such financial relationships. If you choose not to address this issue of financial relationships at the beginning of your statement, however, it will not preclude you from speaking.

DR. FINDER: Okay. Before we get to the public speakers, I want to mention about alternative standards that we have approved since the last meeting.

For those not familiar with this section of the regulations, FDA may approve an alternative to a quality standard that exists under Section 900.12, when the agency determines that, one, the proposed alternative standard will be at least as effective in assuring quality mammography as the standard it proposes to replace; and, two, the proposed alternative is too limited in its applicability to justify an amendment to the standard or offers an expected benefit to human health that is so great that the time required for amending this standard would present an unjustifiable risk to human health, and the granting of the alternative is in keeping with the purpose of Statute 42, USB 263(b).

Since last April's meeting the division has approved two alternative standards. The first deals with the system artifact testing at remote mobile mammography sites where film processing takes place using processors permanently located at that site.

This alternative permits a special trained quality controlled technologist to make system artifact films and phantom images at remote processing sites used by mobile mammography facilities and then submit them to the facility medical physicist for evaluation.

This relieves the facility from the need to have the medical physicist visit each remote processing site as part of the annual survey.

The second deals with system artifact testing of target filter combinations. The approved alternative permits the system artifact tests to be performed without testing all target filter combinations during the annual physics survey. These alternative standards in their entirety are available on our Web site in the policy guidance help system.

If anybody has any questions about these alternatives, I do have copies of the full wording for any of those.

I believe not.

CHAIRPERSON HENDRICKS: At this time we'll move it into the first public session. We will introduce the scheduled public speakers one by one.

Our first speaker under the topic of approved alternative standards is Dr. Carol Lee, and she'll be speaking from ACR.

Dr. Lee.

DR. LEE: I want to thank this committee for allowing me the opportunity to address it. I am representing here the American College of Radiology, which is a 30,000 member professional organization representing diagnostic radiologist, radiation oncologists, and medical physicists.

And I also want to say that my travel expenses have been paid by the American College of Radiology to attend this meeting.

Is there any way that I can advance these? Oh, okay. Could I have the next?

I hope you can read these slides. They're a little busy.

The American College of Radiology has a longstanding record of a commitment to quality in breast imaging. This began in part with a voluntary mammography accreditation program that was begun in 1987 that laid the foundation for subsequent Mammography Quality Standards Act.

In 1993, the breast imaging reporting and database system was developed by the ACR, and provided a lexicon for description of breast findings and a reporting system that helped standardize and clarify mammographic reporting in this country.

In 1996, an accreditation program for stereotactic biopsy was developed. BI-RADS was expanded to include breast ultrasound and breast MRI in the most recent edition published in 2003, and this past year the American College of Radiology established a permanent breast imaging commission as part of its Board of Chancellors, replacing an ad hoc task force to deal with matters relating to breast imaging in this country.

The ACR also has a record of providing educational and self-assessment tools to breast imagers, including a biennial national conference on breast cancer. The ACR has developed a self-assessment tool that's available to the public and also sponsors a regular mammography education program at the Armed Forces Institute of Pathology for radiology residents across the country.

Next slide, please.

Now, in addressing problems with mammography interpretation, the IOM report unfortunately did not really specify what particular problems need to be addressed. It was sort of an overall assumption that there is a problem. Whether or not this is manifest as too many false negatives in mammographic interpretation, a recall rate that's too high or too low, a positive predictive value of biopsies that's too low, too much variability or all of the above was not specifically stated, and it's difficult to know how to develop programs or mandates or regulations to address problems when the problem itself it not specifically stated.

In terms of mammography interpretation, certainly there is room for improvement. Certainly mammographic interpretation is not perfect, but -- next slide, please -- I do want to point out that mammographic interpretation in the United States has been compared unfavorably to that in the United Kingdom. It has been published that the recall rate from screening in this country is double that of the United Kingdom.

However, it should be kept in mind that practice climate, the malpractice situation differs dramatically between the two countries, and in looking at the report comparing U.S. to U.K. mammographic interpretation, there actually are more cancers that were picked up in the United States over the study period. There were 55 cancers per thousand women screened over 20 years compared to 43 in the United Kingdom. And most of the additional cases that were detected in the U.S. were due to small, invasive cancers, and DCIS, which is just the type of tumor that we hope to be able to detect through screening.

Next slide, please.

In addition, there are studies that have shown that the size of tumors within stages has decreased since the advent of modern mammography, and whereas in the period from 1975 to 1979, fewer than ten percent of breast cancers were under one centimeter. One quarter of all localized breast cancers were under a centimeter in the period from 1995 to 1999.

Next slide.

And it's important also to keep in mind that breast cancer mortality in this country has decreased by 25 percent in the past ten years. The decrease in tumor size within stage over the past 30 years accounts for most of the observed improvement in survival in localized breast cancer.

Next slide.

Now, that's a quick summary of the good news. The bad news is that there is an impending manpower crisis in breast imaging in this country. For this past July, only 33 percent of breast imaging fellowship positions within the fellowship match were filled. The proportion of radiology residents who state that they want to spend a significant percentage of their future practices in breast imaging has declined from 29 percent in a study done by Dr. Bassett, who will be addressing this committee later this morning, and in a more recent study of Massachusetts residents, only three percent said that they would like to spend a significant portion of their time doing mammography.

Next slide.

When looking at this study a little bit more in depth, these were 63 senior radiology residents that were surveyed in Massachusetts, and they were asked about their future career plans and about attitudes towards mammography.

Next.

Only eight percent said they wanted to do any mammography in their future jobs, and only three percent stated that they wanted to spend at least 25 percent of their clinical time reading mammograms, and only one of the 63 intended to pursue a breast imaging fellowship.

So it's a bit disheartening, and there's no reason to think that Massachusetts is any different from any other state in this country.

Next.

When asked why they did not want to spend time doing mammography, the majority said that they were afraid of lawsuits and the medical legal climate.

Next.

Now, in talking about, in addressing the IOM regulatory recommendation in the recent report improving mammography quality standards, I want to address some of the requirements specifically.

Could I have the next slide?

One of these recommendations suggested requiring separate tracking of results of screening and diagnostic mammographs in order to be able to compare to established benchmarks.

The problem with this recommendation is that the definition of screening varies among practices and makes comparison among facilities quite difficult. I've recently found -- I've spent the past 20 years doing breast imaging at Yale University in an academic practice, and just this past year I have been working as a breast imager in a private practice in Honolulu, Hawaii, and I can tell you that the practice varies considerably in terms of what is considered a screening mammograph, what is considered a diagnostic mammogram.

Both facilities produced very high quality images, and I think the personnel at both facilities are comparable in terms of their degree of expertise, but I can tell you that the practice patterns vary quite a bit.

Some facilities don't differentiate between screening and diagnostic examinations, and some facilities in the way they handle the examinations handle them differently to make this differentiation difficult.

And I also question the applicability of benchmarks to individual practices. As I was saying, my performance has not changed since I moved to Honolulu, but because the patient population differs so much as a high Asian population, women with very dense breasts are the norm rather than the exception. There is a lower prior probability because the risk of breast cancer in this population is inherently lower, and my performance hasn't changed, but my benchmarks have. My recall rate is higher. My positive predictive value is lower. So I really question the applicability.

Next slide, please.

In terms of required tracking of outcome of all cases with BI-RADS 0 assessment, this is not easily achieved. It's quite difficult to achieve this even with commercially available software tracking programs, and again, I have experience with two different, very widespread, widely utilized tracking programs, and this is a difficult audit to achieve with both of them. This requires a substantial increase in time and effort and expense, and we already track the BI-RADS, four and five cases that come out of the BI-RADS 0.

As stated in the IOM report itself, there has been no provide benefit to this additional tracking.

Next slide.

The inclusion of interventional mammographic procedures, specifically stereotactic biopsy, in the MQSA, we believe, would be justified because we do think that this would lead to quality in these procedures, and there is a stereotactic program again sponsored by the ACR that is in place.

Next slide, please.

Regulation of breast ultrasound and breast MRI are also mandates that are likely to result in improved technical quality. Unfortunately this is not yet feasible for breast MRI because the accreditation program has not been established, and also, the hardware and software is still in development in terms of technique for breast MRI.

Next please.

So increased regulation, particularly if it's unfunded, runs the risk of decreasing access through worsening manpower shortages and increased facility closure. I was recently at an ACR meeting where several of the attendees stated that their practices were considering dropping mammography services.

the goals of improvement inequality should be clearly understood, and new regulations should have a high likelihood of improving these targeted quality parameters.

Next please. Can you switch? In the interest of time, next slide.

So in conclusion, the ACR has a proven commitment to quality improvement that has been demonstrated over the years, and certainly these efforts should continue. The increased mandatory

auditing requirements have not been shown to translate into improved quality that would justify the quite substantial commitment of time, effort and expense involved, and this that I'm referring to specifically is separating the screening and diagnostic auditing and the tracking of the BI-RAD 0 cases.

We do feel that stereotactic breast biopsy and accreditation of breast ultrasound is likely to result in improvement in quality. It's premature to require regulation of breast MRI at this time.

Thank you very much.

CHAIRPERSON HENDRICKS: Thank you, Dr. Lee.

Are there any questions?

(No response.)

CHAIRPERSON HENDRICKS: If not, we'll move then to our second speaker in this open public hearing. We welcome Dr. Larry Bassett to the podium from the Society of Breast Imaging.

Dr. Bassett.

DR. BASSETT: Thank you.

The Society of Breast Imaging is the largest national organization specifically committed to the practice of breast imaging, and I can refer you to the SBI Web site for position papers on the IOM report. Simply www.sbi-online.org.

Next.

We are addressing now the issue of improving breast imaging quality standards from the Institute of Medicine published 2005. The Society of Breast Imaging commends the thorough data finding efforts and analysis by the IOM in defining many issues that are confronting breast imaging practices. However, some proposed solutions may have a negative impact on the goals sought by both the IOM and society at large.

Here I've kind of outlined the four main categories of the recommendations to improve breast imaging quality taken directly from the IOM report. One is to improve mammography interpretation; the second, to revise MQSA regulations and inspections and enforcement; and next to insure an adequate work force by breast cancer screening and diagnosis; and the fourth, to improve breast imaging quality beyond mammography.

Next.

So starting with the first of those, improved mammography interpretation. This addresses, first, to revise and standardize the required medical audit component of MQSA and, two, to facilitate a voluntary advanced medical audit with feedback.

The SBI has found a concern about this in that increased regulations, while aimed to improve breast health care, have to deal with also the work force shortages and low reimbursement that will be aggravated by implementation of such measures.

Dr. Lee has indicated the crisis in the work force for breast imaging. This is something that has been identified and reported in at least three papers in the literature in peer reviewed journals.

Why is this happening? I think we heard that there's not new people coming into the field. There are people dropping out of the field and people retiring, and some practices simply do not find it financially feasible to have a strong breast imaging component.

We do feel that if we're going to increase regulations we need to have sufficient incentives to create an infrastructure that will support the improved delivery of care before these go into implementation rather than have them occur concurrently with the imposition of additional regulatory burden on a work force that's approaching crisis level shortages.

We see this every day in my practice. We train many breast imaging fellows. We've trained 64 to date sine we started that endeavor. I had two last year. Both of them had about ten job offers in the first month of their fellowship. There's a real need out there, and it's not being filled.

Next.

We are concerned that increased medical audit requirements may scare off other current breast imagers.

Next.

The new audit recommendations are time consuming and require more paper work. No payment is included for this, although to date the facilities have used their own resources to pay for all of the requirements that they have to fulfill. This is another one that may just be the one that broke the camel's back.

In our own facility in order to try to do this kind of medical audit which we have been doing, but it's an academic institution, we had to hire a QA coordinator who keeps track of the zeros and so on and the fours and fives.

We couldn't do it as a radiologist because we're already working the work of two people because we're short staffed. We can't get someone to take the open position that we have in breast imaging, and this is true of most facilities that are academic facilities in the United States.

Small, rural facilities may not have the same kind of resources, may not be able to accomplish this. And the other thing that Dr. Lee referred to is that we know that the results of medical audits depend largely on your patient population. So we're not sure there's a real standard that we can arrive at that will be uniform across all practices.

Next.

In addition, there was a recommendation to establish breast imaging centers of excellence and undertake demonstration projects and evaluations within them, and further study of the effects of CME, reader volume, double reading and computer aided diagnosis and detection.

The Society of Breast Imaging supports the concept of centers of excellence and thinks this should be pursued.

In terms of the other recommendation, there's a lack of evidence that variables such as reader volume are related to interpretation quality and requiring greater volume would further reduce the number of physicians that are qualified to interpret mammograms.

Next.

The next issue was modifying regulations to clarify their intent and address current technology and to streamline inspections and strengthen enforcement for patient protection. And the Society of Breast Imaging agrees requirements should be reviewed to determine which are the effective ones in determining the quality of mammography and perhaps remove and eliminate those that are not effective.

And the SBI strongly supports streamlining the process.

Next.

In the next group to college and analyze data on mammography work force and service capability, to devise strategies to recruit and retain highly skilled breast imaging professionals, and to make more effective use of breast imaging specialists, the Society of Breast Imaging supports any way to reduce burden on the current breast imaging work force.

However, the use of radiologists to assist us to interpret breast images is controversial, remains to be proven in effectiveness, and does not reduce the medical legal responsibility of the interpreting physician who is overseeing them.

We do want to improve output. We want to do that without sacrificing quality.

Next.

Finally, to mandate accreditation for non-mammography breast imaging methods that are routinely used for breast cancer detection and diagnosis, such as ultrasound and breast MRI.

Also, we want to include the issue that as an X-ray examination of the breast, stereotactic biopsies should by law be included in MQSA regulations.

And as was mentioned, there's already an accreditation program that's set up. So this would be an easy one to implement.

High quality ultrasound is crucial in breast imaging today, but there is variable performance and equipment, and this also merits mandated accreditation, and again, there is an accreditation program already in process so that it could be easily adapted.

And, finally, breast MRI accreditation we feel should occur, but this will have to come later because we don't know what the proper standards are yet today.

Thank you.

One more slide.

Thanks for your time and service to the field of breast imaging. We appreciate this very much from the Society of Breast Imaging.

Thank you very much.

CHAIRPERSON HENDRICKS: Any questions related to Dr. Bassett's presentation from the audience or the panel?

(No response.)

CHAIRPERSON HENDRICKS: Then at this time we'll ask Dr. Finder to read two separate sets of written comments from public speakers who are not present today. The first is a set of anonymous written comments related to the IOM recommendations.

the second is a set of written comments submitted by Dr. Richard Ellis related to the IOM recommendations.

Dr. Finder.

DR. FINDER: For those in the audience, these comments are a part of the packet of materials that you have.

The first one, as mentioned, was sent in by a person who wishes to remain anonymous and wanted this read into the record. It says, "I am" -- and I'll try and go through as much as I can in ten minutes. It's a rather long statement here.

Okay. "I am a registered nurse of a woman's diagnostic center. As a patient advocate, I am aware of the advocacy of other nurses, such as Judy Wagner. I have some great concerns about the practice of mammography and other breast imaging modalities if we do not make some changes in educational requirements and tracking of competence.

"Screening mammography still provides the best defense against a death from breast cancer. However, this is only true when the quality of radiologists' reading accuracy is highly proficient.

"As you may or may not know, studies indicate that doctors need to read minimally 2,500 films each year to stay sharp. The government, however, only requires 480 per year. It is ridiculous to think that anyone can be proficient reading this few per year.

"At this time the government does not require regulatory agencies to monitor these levels of proficiency. Having the radiology groups monitor internal performance is quite like having the fox watching the henhouse. Most radiology department directors have neither the time nor the staff to go back and find the false negatives.

"A run a small center, and makes it a priority to go back with each new cancer diagnosis and see if it fits into the false negative category. This is time consuming, but very necessary.

"To date I have discovered 18 false negatives read primarily by two radiologists within the past two years. I work closely with the medical director of the center who has been very supportive. He has helped me get this information to the Physician Quality Committee and has made recommendations to remedy the situation.

"While we have discovered the actual cases of missed diagnosis, I can only wonder how many more patients that were given a benign or negative outcome may have a cancer already.

"Several months ago we had a fellow trained mammographer join the practice. She quickly spotted the above problem and stated that there were two other physicians in the group who should not even read mammography. She stated that this was more of a problem than just the need for a few CMEs in mammography. The cancers that were being missed were, for the most part, not small, difficult to see cancers. It seemed clear to her that there was a real problem identifying what cancer looked like in its early stages.

"We have the R-2 image checker which is used, and still we have this many missed diagnoses. She left the group after only a few months to go to practice in a large hospital breast center. As with the majority of community hospitals, the radiology group has the hospital radiology contract. They are very good at many things. They rotate several physicians through our center to cover mammography. They do not enjoy reading mammography, and clearly, are not going to spend time and effort to even go to a visiting fellowship in mammography as recommended for correction of deficit.

"My problem is that once you have a really good mammographer and see what quality looks like, you can't go back. I have insisted that they find us another mammographer, and so far the administration has backed me up. They, of course, are dragging their feet because this new person will not generate the same amount of revenue that other physicians in the group generate.

"My feeling is that there needs to be a new paradigm in the way radiology groups think about practice recruitment and development. Because the average radiologist would prefer not to do mammography and other women's imaging, the group should be willing to subsidize salaries for those who are willing to do this kind of practice.

"Women deserve this vital service even if reimbursement is terrible. It should be of some value to a radiology group to have one well trained, passionate person take all the heat in this highly sued specialty. This mammographer could help raise the bar for all radiologists in the practice.

"Mammographers should be allowed to just do mammography and should not have to keep skills up in other areas as well. Breast imaging is changing so rapidly that it is no longer just mammography. A breast imaging specialist has to be able to read breast MRI, do minimally invasive breast biopsy procedures, talk to patients and the public in general. He or she should not be expected to take general radiology call as well.

"When I confront our radiology group with their individual statistics for all BI-RADS categories, false positive, false negative, true positive, true negative, and when I provide percent recommendations, they tell me that they should not have to be held to the standards. They say they don't read as many per year as mammography experts and can't be expected to reach the same level of proficiency.

"I say this is bunk. If I go to a surgeon and have my colon removed, should he be able to say to me, 'Well, I missed some of the possible cancer because I don't do as many of these as some others do'? May it never be.

"At the last NCBC conference this past February I voiced my desire to see minimal reading and CME standards improved.

"One, additionally we must improve reimbursement for breast imaging. We must find a way to provide incentives for bright, dedicated physicians to go into breast imaging.

"Two, we should encourage radiology groups to recruit breast images and be willing to subsidize their salaries.

"Three, regulatory agencies must find a way to do more than measure accuracy of equipment in their surveys until such time as physicians can adequately police themselves. In lieu of this, hospitals should be required to have non-physician personnel or consulting physician personnel monitor statistics for reading accuracy.

"Four, the MQSA needs to become more comprehensive. I am in favor of expanding it to Breast Imaging Quality Standards Act.

"Six, with regard to stereotaxic qualifications, I have no problem with surgeons doing stereotaxis in our center. If the radiologist has done a good job of marking the recommended area for biopsy, our technologists have no problem locating the lesion on the computer and preparing everything for the surgeon. He then reviews the mammogram. The stereo is set up and marks the area for biopsy.

"I think when surgeons get into trouble doing stereotaxis is when they interpret the mammograph which was done perhaps in their office, then expect an aide other than a registered mammo technologist to set up the equipment and position the patient.

"Recently I read in the Mammography Regulation and Reimbursement Report that the American College of Radiology would begin calling the false negative a sentinel event for the hospital. This would have a big impact on hospital accreditation.

"I will now be attempting to track our cancer patients to see if they die of breast cancer. If so, the sentinel event repercussions for the hospital are significant. I can see if this happens, the individual hospitals will begin demanding better quality readers of breast imaging.

"I am only one voice, but I am a thorn under the saddle. Each time I find another false negative, I see the patient's face. I am the person who will counsel them prior to their first surgical visit. I am the nurse who gives out her phone number to them for questions and comfort. I am the nurse who runs the women's cancer support group. None of them know that their cancer should or could have been caught earlier.

"It is my job to market our center as a center of excellence. We meet that goal in every single way. We have very high customer service scores, and people rave about the quick and compassionate service they receive. Indeed, we are a center of excellence in so many ways. It is the physician component that lets us down.

"In order to keep my job, I must fight this battle quietly within the Physician Quality Improvement Committee. It has been of little value to me. Recommendations for improvement are just that. The radiology group has little incentive other than my constant nagging to do much about anything."

And then she lists some percentages and indices for evaluation. I will say that this person reported that she had no financial interest, conflict of interest.

CHAIRPERSON HENDRICKS: Any questions or comments related to the anonymous statement?

DR. FINDER: Okay.

CHAIRPERSON HENDRICKS: Next we move to the comments from Dr. Richard Ellis.

DR. FINDER: Dr. Ellis is from the Gunderson Lutheran Medical Center. He also reported that he had no financial conflict of interest to report. We'll give him the full ten minutes.

He says, "I appreciate the opportunity to submit a statement for review and consideration by the FDA concerning the Institute of Medicine Committee's recommendations for improving MQSA.

"For over nine years I have practiced as a clinical breast radiologist, subspecializing in the early detection and diagnosis of breast diseases. Listed below are the issues that need to be reviewed and addressed both by the FDA and IOM committees as you prepare recommendations for reauthorization of MQSA.

"If the intent of screening mammography is to reduce the mortality and morbidity of breast cancer, then early interruption of the disease is paramount. Over the past 100 years we have seen advances in surgical techniques that have significantly improved patient morbidity but not mortality. Likewise, we now have chemo and hormonal therapies that have allowed moderate improvement in patient mortality.

"However, it is the advent of early detection and diagnosis which interrupts breast cancer early in its natural history that has resulted in the greatest reduction in mortality from breast cancer.

"Since the initiation of MQSA we have seen improvements in the technical aspects of screening mammography given the standards required for certification. However, even if we have the best equipment, X-ray film screen systems, technologists, quality assurance programs, and viewing conditions, but lack high quality interpretation of the screening mammograms, we severely limit the potential of early detection.

"This is clearly demonstrated in the randomized clinical trials for screening mammography as tumor size and stage at detection drives subsequent mortality rates. In order to insure high quality interpretation, a performance audit must be obtained, reviewed, and action taken when deficiencies are noted. A screening mammography interpretation performance audit should include one average size, mean and median size of the screen detected invasive carcinoma for women participating in 12 or 24 months screening intervals; two, total screening volume per year; three, recall rate; and four, positive predictive value for BI-RADS 4 and 5 categories. In order to help achieve acceptable screening performance standards, radiology residency, and breast imaging fellowship training as well as postgraduate training programs that properly instruct high quality screening interpretations are critical.

"Although there are many mammography/breast imaging programs offered for CME each year, unfortunately many simply do not provide the type of education that will allow direct improvement in screening mammography interpretive skills.

"Two, many will argue that if physician performance standards are set to insure a high standard of care, then access to women in many communities will be lost as many general radiologists may not be able to achieve and/or maintain the required standards.

"This issue can and has been successfully addressed by other countries including Sweden and the United Kingdom. Although the total number of screening mammograms interpreted per year may serve as a surrogate performance marker, Items 1, 3, and 4 listed above provide an objective measure of performance. If inappropriate low interpreting physician performance standards are set by MQSA to simply afford greater access, mortality rates will likely not be reduced, and overall cost of care will increase.

"Reasonable and preferable high interpreting physician performance standards need to be instituted. An exemplary model of postgraduate training success can be found by examining the interpreting physician performance improvements achieved by private practice, radiologists in Albuquerque," and he talks about Linver's practice there.

"Similar models of postgraduate training with proven success need to become a fundamental part of physician CME for screening mammography breast imaging.

"On a similar issue, communities and medical institutions of sufficient size should strive toward creating interdisciplinary breast care teams which help provide improved overall care, efficient use of resources, and substantial reduction in medical costs.

"Third, with the use of screening mammography, the majority of breast cancers are initially detected in the preclinical phase, nonpalpable. However, other imaging modalities, especially breast ultrasound, are frequently used in the diagnostic evaluation of screened detected abnormalities to further segregate which patients may require a biopsy for definitive tissue diagnosis.

"Given the advancement in training and technology, most breast biopsies can be performed under image guidance to include ultrasound and stereotactic guided breast biopsies. In 1996, through the joint efforts of the American College of Radiology and the American College of Surgery, we have the stereotactic guided breast biopsy accreditation, but which remains under voluntary accreditation.

"However, breast ultrasound and ultrasound guided breast biopsies have multiple guidelines and accreditations from various institutions and agencies, to include the ACR, the ACS, the American Society of Breast Surgeons, and the American Institute of Ultrasound in Medicine.

"Both the FDA and IOM members need to investigate why they're on multiple and varied physician training guidelines and accreditations for breast ultrasound examinations and procedures. The FDA and IOM should insist on a single universal standard for ultrasound examination and procedure training guidelines and accreditations so that multiple standards are not propagated.

"Even amongst radiologists there is a wide disparity of performance and interpretation for breast ultrasound. The universal accreditation program will help insure that not only mammography, but also other breast imaging examinations and procedures meet an acceptable MQSA standard for accreditation. If the FDA and/or IOM through the MQSA does not require and enforce the universal practice standard for breast ultrasound examinations and procedures, then the qualification for breast ultrasound will simply fall to whomever can afford the equipment regardless of prior training and performance level.

"In the very near future universal standards and accreditation should also be established for breast MRI and imaging guided breast tumor oblation.

"Four, although not directly related to MQSA, a major issue to be considered is the dwindling number of radiologists and training residents that have a desire to provide mammography/breast imaging services. Relatively low reimbursement rates and high exposure to malpractice litigation must be addressed and appropriate incentives need to be provided to prevent radiologists from abandoning mammography/breast imaging services. Creative solutions, to include providing graduated reimbursement rates for mammography/breast imaging services based on physician performance and creating a balanced, knowledgeable national committee to review and arbitrate medical malpractice suits, along with placing caps on punitive damages, tort reform will be important.

"Should you have any questions or need additional information, please contact me. I appreciate your review and consideration of my recommendations."

CHAIRPERSON HENDRICKS: That ends the submitted comments from the public speakers for this portion of the session.

Based on the itinerary, I'd move to a break, but may be we should move into the next speaker because we are ahead of schedule.

The next item on the agenda are some comments to be made by Dr. Michael Divine, who is Chief of Inspection and Compliance Branch, to comment on inspection observations.

Mr. Divine, welcome.

DR. DIVINE: My name is Michael Divine. I'm the Chief of the Inspection Compliance Branch, Division of Mammography, Quality and Radiation Programs.

There will be two main topics for this particular discussion. One will be similar to the inspection results from the MQSA inspections for the last three fiscal years. The fiscal year for FDA runs from October 1st to September 30th, and I'll also talk about from follow-up actions involved regarding things we can do when we find serious problems.

Okay. The inspection results I'm going to be discussing for this fiscal year, which started October 1st ran through August 26th of this year. All of our inspection observations from inspections are broken down into three different levels. The first is level one, which we consider the most serious. This is the most likely one that might result in us taking regulatory action or conducting a follow-up inspection or warning the facility if they don't have a history of problems.

The next level which we consider moderate but still significant is level two, and the last one is level three, which we consider minor. No significant problems.

This is probably the most important slide I'll give for this presentation because it shows the overall performance of facilities over time. If you looked at a chart like this spreading back to 1995 when we started inspections, you would also see that it has been continuously improving since the beginning of the program.

The level one observations has been very, very small, for several years now almost nonexistent on this slide. You also see a drop in the level two and the level three problems over time, and you're seeing that the top line, the blue line is where things have gotten to we're about at 70 percent of facilities that have a clean inspection.

Starting with some level one problems for initial qualification of personnel, the first set of bars has to do with the physician either having board certification or the alternative of having two or three months of training in mammography. This number has been dropping.

To give you an example of the perspective, we do about 9,000 inspections. So this is way less than five percent of facilities that have this problem. The license problem, we still see some of that. Mostly we think that's an issue of allowing the license to expire and not getting it renewed, which is mostly a technical problem and not really related to quality, or that they don't have any documentation at the facility during the inspection.

For the medical physicist, pretty much has just gone away. We don't see too many problems with the physicists at all anymore. We still see some problems with the technologists. Once again, this is a very, very small number of facilities considering the number we inspect every year, and that has also been dropping.

In the QC area, once again, we have some problems, but they're not that significant in terms of overall numbers. We have an observation for if they have failed to do processor QC for five consecutive days or more, that will get them a level one, and that's around 30 to 40 facilities. Processing out of limits, when they're outside the actual limits on their processor charts.

The third category is where we look at the number of days in a month, the percentage. That is also very small, and the last column is when there's standard QC missing.

The first few charts on this, these are tests that are done during the inspection. All of the ones that are very low are relating to test inspectors. As you can see, it's much less than 20 facilities for any of those phantom tests, either the spec groups, the fibers or the masses that are broken down individually. So there's a very small number of problems that we're finding with phantom image testing.

Under processing, which is a test we do, it's called the step test, which is a test of sensitometry where we actually run a film through the processor and compare it against a standard. We find very few problems with that these days. The processing has gotten much better.

We do find a certain number of problems with fog when we go in to test the fog in the dark room, but once again, we're talking about much less than five percent of facilities.

These are problems relating to the survey. The first column is where there's more than 14 months between the annual survey. Even though they're required to have an annual survey, we allow up to 14 months between surveys before we consider it to be a problem.

The next column is when we go into the facility, and the last survey was more than 14 months, but they haven't had a more recent survey during the inspection. So it's a much bigger problem when we go in because of the timing. We usually see the last survey and the one before that being more than 14 months.

The next set of columns has to do with an incomplete survey where there's either a problem with the tests that were done or there were missing tests from the survey. So the survey was done, but there was a problem with it.

The next column has to do with the X-ray unit. When they installed a new X-ray unit or they have a major repair on the unit, they have to have mammography equipment evaluation done by a medical physicist, and this column has to do with -- these were not done -- once again, this is a very small number because if it's a unit, they have to go through accreditation. So they're going to have to have it done anyway. So this is a fairly rare occurrence.

The next one is where they install a new processor or they have a major repair on the processor. Once again, they have to have the medical physicist come in and do an evaluation, and we're seeing very few problems in that area.

These are testing that are done by the inspector during the tests, and I added this slide for comparison purposes because we're going to be talking about what things we do during inspections and streamlining inspections during the next discussions.

The first column is zero, and there's an asterisk there to point out that we have not seen a dose value exceeding 300 millirad in an inspection since 1997. It has basically gone away as a problem.

So we're still doing the testing, but we haven't found any problem since 1997. Exposure reproducibility, almost nonexistent. It looks like about ten facilities out of about 9,000 inspections. This is a test to see that shooting the X-ray beam several times with the phantom in the beam produces the same level of radiation. Beam quality has basically gone away as a problem.

We still see a certain number of facilities with the alignment tests, but this involves several different tests we do in the inspection. One is oversizing of the X-ray beam on the film. It also involves where the compression paddle is in relation to the film and the chest wall of the X-ray. So there are several different problems that can get a facility here. So it's not just one test.

But even taking that into consideration, there's still a small number of facilities, less than five percent.

Getting into some interpreting physician qualifications, radiologist qualifications, and level two, one is the initial CME. That's either having 40 hours of training in mammography or 60, depending upon why the physician qualified. Once again, that's a small problem and has been decreasing. The initial experience is the 240 mammograms read within a six month period. That's very small.

We see more problems with continuing experience, continuing education, but those numbers have been going down over time, and as you can see, the continuing education has been dropping, too.

Technologists qualifications. The mammography training is that they have to have 40 hours of training in mammography with supervised examinations. Once again, we're almost seeing no problems with that. As with the physicians, you're still seeing some problems with continuing experience. The continuing experience is to have read 200 mammograms in a 24-month period, and the continuing education in a similar position. I still see some problems, but they are decreasing.

Medical physicist, initial training or initial experience requirements. Once again, we're seeing almost no problems at all in that area. Also having some problems with the continuing education and experience, but as you can see, it's almost nonexistent compared with 9,000 inspections a year.

Getting onto medical records and reports, we have a problem. The facility has a problem with sending out patient letters or mammography reports within 30 days. They can get a level one for that. We consider that a very serious problem. Once again, this is just a small number of facilities. Way less than five percent of facilities get this problem. So it's pretty much a non-problem these days.

We do see some problems with the assessment categories. A lot of that has to do with wording. We're still seeing problems with that, but we don't consider that to be a significant problem, but it is decreasing, and reports without the physician being identified, that's almost gone away altogether.

the first one has to do with the X-ray unit. When we first went in with the final regs. in 1999, requirements that, for instance, they have to have two film sizes for each mammography unit to do the 24 by 30 and the 18 by 24. Most of those problems have gone away, almost nonexistent problems these days.

We still see some problems, though very minor, with the procedure for consumer complaints, and that has been going down, I think, as the facilities get better educated as to exactly what we're expecting, and the procedure for infection control, once again, that's a small number of facilities. Once these procedures are in place we usually don't see it from year to year. So a lot of these could be facilities that are new or have changed their procedures and we find problems with them.

This has to do with the medical outcomes audit. The first one has to do with not all of the positive mammograms are entered into the system. Once again, this is a very small problem.

The second one, even though it says no biopsy results, what we're actually talking about is they haven't gotten all of the biopsy results for all of their positive mammograms or they can't document that they have made an honest attempt to get all of those results.

The third one has to do with they have not identified an audit interpreting physician for -- this is an annual audit. The third one is, which we've seen more problems than the last three categories, but still very small. The analysis is not done annually. They're required to do an annual analysis of the results.

The last one is that they haven't broken down the analyses by each interpreting physician, which they're required to do, an the last one is that they haven't done an analysis for the entire facility. They may have broken it down by each physician, but they don't have a complete analysis for the whole facility or they haven't done it at all.

This one has to do with just if they have any problems, and if you go back to all of those other problems where we talked about qualifications of personnel, if they have any problem, they get cited for this, and there's also a situation where -- and it's a good time to mention this -- if they go in and there's something missing from the file, but the facility can justify that the documentation exists and they can obtain it within five days after the inspection and they can provide that to the inspector, either fax it to the inspector or get it to the inspector before they send the inspection to us. We allow the inspector to remove that observation from the inspection, but they will still get cited for this thing because they have to have their documentation, their paper work ready for the inspection.

So we consider it a problem, and we want to track these problems. Of all the things I've show today, this is going to be the highest because, you know, there's always going to be something at a lot of facilities that they're going to have a problem somewhere, and so this is still less than about ten percent of facilities, but still more. And once again, it's still been something that is decreasing.

Okay. The next part of the talk is I'm going to be discussing actions that we can take after inspections and for facilities that have ongoing problems, once again, most of what you saw in the previous slides have to do with problems that aren't going to result in these kind of things. It's just a very, very small number of facilities that have problems over and over again, and that we've decided we've warned them and they still have problems.

The types of things that we can do, the first thing we would probably consider is a follow-up inspection, and the follow-up inspection would be done, let's say, for a level one problem at a facility. Let's say the facility had responded to the level one. Well, we knew the facility had some problems in the past, and even though they're telling us what they're going to do to correct the problem, because of their track record, we want to go back in there to see that everything has been corrected permanently. So we might go in. Since we're doing an annual inspection, we might go in about midyear, about six months and go in and check to see if everything is okay.

Another thing that we can do which has to do with evaluating whether the problems at a facility could affect clinical image quality is we can do additional mammography review. In the vast majority of situations the facility's accreditation body would be done the additional mammography review, but our regulations allow for us to have somebody other than the accreditation body do the review, but that would be under very unusual circumstances where that would occur.

Should the results come back from that additional mammography review that the patient's facility's mammography quality represented a serious risk to human health, we have the authority to require patient and physician notification about those problems so that the patients are aware that there's a problem or a potential problem with their mammogram.

Another thing we can do if we have a facility that has a lot of problems is something called a directed plan of correction. This is a list of conditions in addition to what's in the regulations for them to operate. This is something we can do instead of shutting the facility down. We think we can work with the facility. It requires a lot of monitoring.

For instance, a typical thing they might require would be that they would have to send in records to FDA, let's say, on a monthly basis, for instance, quality control records or any other things that we think requires the requirements to be put in the facility so we are assured that they are operating in compliance, and it also could involve additional inspections where we go into the facility and requiring them to come up with more detailed procedures than would be required of the regulations so that we know that they're keeping track of things.

Civil money penalties is pretty self-explanatory. This involves fines that we can levy against a facility that is in violation.

The next one is if these other things have not worked, if we've tried to work the facility. We've given them warnings. We've even put them under a directed plan of correction or we believe that they have a serious risk to human health, let's say, from an AMR, additional mammography review.

We have the option of suspending their certificate. If we suspend their certificate, they have to stop doing mammography until we lift the suspension, and another option, which we have yet used, is revocation. This would be a much more serious version of suspension. They would have to stop doing mammography, but the owner-operator of the facility could not own or operate a facility for two years if this occurred.

The last one is injunction. This is the only one on this list that would actually go to federal court. We consider this somewhat of a last resort because we have all of these other tools. We usually don't have to go to court. We can deal with facilities in that, but up to this point we have not used injunction.

Injunction basically is a court order that would shut the facility down. There are only two conditions under which we can use this. One is that the facility's mammography is a serious risk to human health. Since we usually suspend the certificate if we find that to shut the facility down, we would only go to court if they continued to do mammography after suspension.

Another one is if they were performing without a certificate. Once again, usually when we find a facility that's performing mammography without a certificate, after talking to them they usually shut down until they can get reinstated or apply to an accreditation body to get a certificate. So most of those are not real problems, but if a facility just decided they were going to continue without a certificate even after warnings, we would have to go to court to shut them down.

Follow-up inspections. We checked on corrective actions for serious problems. Usually it's a level one, though it could be repeat level two. Usually if a facility has recent problems, if they haven't had problems in the past, we probably will at least evaluate their response, and if it looks adequate we probably won't need to do a follow-up inspection. So it's usually a facility that has a level one or a repeat level two, but now they have more problems in the past. So we can't really take their word for it as far as their corrective action, and it's usually limited to certain specific problems because something we believe we can monitor without actually having to go in the facility.

Additional mammography view, as I mentioned, it's usually done by the accreditation body. It can be anything from two mammograms all the way up to 30 mammograms, and if there's a serious risk found in the review, we would require patient and physician notification.

Some examples, if we find a level one phantom image failure at inspection, we do a limited review, usually two mammograms to check everything is okay with the clinical quality. The level one for the interpreting physician, we theoretically could do that; in most cases is related to qualifications. We usually are able to resolve those problems without having to do an additional mammography review, and in many cases it's more practical. If they can't document that they meet their initial qualifications at a level one, we may actually have the facility, have another qualified interpreting physician rereview all of the mammograms read by that interpreting physician. So sometimes that works better than having to do an AMR, and that assures that all of the mammograms are read by a qualified interpreting physician.

If we have problems, we from time to time have complaints about clinical image quality that need to be investigated, and really the only way to do that is to have the accreditation body look at clinical images to assure how bad the problem is or if there is a problem.

Overall, quality assurance failures, generally when we find a lot of problems in that area, we're usually taking some other action, and because we found these problems, we have to have some assurance that they haven't affected clinical image quality. So we usually do additional mammography review in those cases.

And with fraudulent record keeping, which relates to some of these other areas, we usually do -- we have to be able to assure that the clinical image quality has not been affected by the fraud.

Once again, if the AMR shows a problem that's a serious risk to human health, we require patient and physician notification. This provides the patients and the physicians an explanation of the problems that were found, how they were found, and some follow-up actions that the patient may wish to have another mammogram. The patient may wish to have another physician evaluate their mammogram to see if their mammogram is bad enough that they do need another mammogram.

And we try to use plain language as much as possible. In the early days we did some focus testing and found that we have to make sure that everybody understands what's being included in the letter, and we try to make it as readable level for all possible patients that could be notified.

Directed correction imposes additional requirements in the facility and allows us to monitor what's going on in terms of them sending letter into us, and it could include additional inspections.

We only use suspension for very serious violations. Under the act we can suspend for a variety of serious violations. However, we usually have to give the facility a hearing in advance for them to contest our intention for suspension. So they have the option for a hearing.

We usually do this when we've tried to work with the facility. Usually if it was rated with the quality assurance program, we would usually put them under a directed plan of correction before going to this or we would use, you know, some other method before threatening to close them down.

And if we find a health hazard which is a serious risk to human health, usually found through additional mammography review, the law allows us to shut them down immediately, and that's usually our standard procedure for doing that. If we find a serious risk, they're shut down immediately, though they do have the option for a hearing after the fact if they want to contest the fact that we shut them down.

Just to give you some idea of the numbers of actions we have taken over the course of the program, this goes back to 1994 when MQSA started. Sixty-four additional mammography reviews, 17 patient and physician notifications, four directed plans of corrections, three civil money penalties. The six suspensions include we have another option under the law that if the accreditation body revokes the accreditation of the facility, the facility certificate will remain in effect until FDA decides that it should not remain in effect because of the problems that were found.

In many cases we have taken actions directly from a serious risk to human health finding, an additional mammography review, which also resulted in the accreditation body revoking the accreditation of the facility, and then we remove their certificate pretty much immediately.

As I mentioned in my earlier talk or earlier slides, revocations and injunctions have yet to be used.

CHAIRPERSON HENDRICKS: Thank you very much.

Any questions for Mr. Divine? I had one.

DR. DIVINE: Yes.

CHAIRPERSON HENDRICKS: Carolyn Hendricks, panel Chair.

In your opinion, which aspects of the current routine facility inspections could be completely eliminated without impacting the quality of the inspections that are currently being performed?

DR. DIVINE: Well, one of the things that we have been considering removing because, as I mentioned, we do a dose tester in each inspection on each X-ray unit, and we haven't found any problems since 1997. So it's very hard to justify doing that test every year if we don't find any problems every year. That would be one I would mention.

And if we eliminate that, we would also probably be eliminating the reproducibility test because that's all done in conjunction with it, and the beam quality test also because we find very few problems, but those are the things that stick out in my mind.

CHAIRPERSON HENDRICKS: Questions? Mr. Passetti.

MR. PASSETTI: Bill Passetti.

You mentioned that you haven't seen anything exceeding the dose limits and how long. do you know what the current average is that you're seeing throughout the facilities?

DR. DIVINE: I think it's about 1.7 milligray, which is about 170 millirad. That's my -- 1.7, 1.8 I think. It's been going up a little, but that doesn't result in any problems with the dose testing.

I think the reason it has been going up is that there has been a preference for darker mammograms, and usually that's achieved through using a little higher exposure to the patient, but that hasn't resulted in any noncompliances by going up in the last few years.

CHAIRPERSON HENDRICKS: I have a follow-up question. Would those changes that you propose related to elimination of those steps in the inspection significantly reduce the inspection time or expense of the facilities in your estimation?

DR. DIVINE: The inspection time depending upon how many units at the facility and the particular inspector. I think if we eliminated the radiation exposure test, that might reduce the inspection time maybe half an hour per unit. That's just a guess off the top of my head. I don't know how much that would affect the fee. I couldn't really comment on that. It certainly would be something we could consider, but I don't know. I don't have any data on that.

CHAIRPERSON HENDRICKS: Yes.

DR. WILLIAMS: This is Mark Williams, University of Virginia.

Just a follow-up comment on the question of dose. I wonder if it wouldn't be interesting to look at not just the average in the recorded doses during inspection, but also look at the dispersion around the average to see what kind of spreads they are, maybe in conjunction with data from the ACR to see whether or not it might be useful considering a different upper limit. If there is a fair amount of spread around that average, then it may be that even though we don't see violations above 300, it may be useful to consider other thresholds.

DR. DIVINE: Yeah, one thing I would mention, since you brought that up, is that there's an article. I think it's still available on our Web site where we have a spread of the dose data that we found during inspections, and so we do have that available.

One thing I will mention is that we're going to be looking into this issue, and we're going to be recording the dose values that are found by the medical physicists during the annual survey to compare against the values that we're finding during inspection so that we have some idea of how close we are to that and also, you know, if we're finding any problems. We're looking to that also

DR. FINDER: Dr. Finder.

I just wanted to add that in addition to the fact that currently we aren't measuring the dose every year, the medical physicist measures it every year at the facility and the accreditation body measures it every three years.

CHAIRPERSON HENDRICKS: Other questions from the panel, the audience?

I have another question. Related to your slide with the very high proportion of facilities that do not have any violations or any findings a tall, that would be a global lack of any findings in their audit, including the other CME documentation, requirements with --

DR. DIVINE: Yes, they get an inspection report that says all items in compliance.

CHAIRPERSON HENDRICKS: And that represents just about 70 percent of all the facilities?

DR. DIVINE: Yes. It has been continually increasing over the course of the program.

CHAIRPERSON HENDRICKS: So in your view then, would those facilities then benefit from less frequent screening if it's those facilities which are operating at such high quality levels that it might be okay for them to be screened at a less frequent interval and still maintain that high quality of care as determined by the inspections?

DR. DIVINE: Yeah, that's a possibility. One thing we discussed at the last meeting was the results of the inspection demonstration program and what we found was that there seemed to be an increase in the number of problems when facilities skipped an inspection, but you know, there were problems with that study, but that's what we had found.

Yeah, we are open to suggestions on that.

DR. FINDER: Yeah, this is Dr. Finder.

I just wanted to enhance what Mike said about this. For those new members on the committee, in the last reauthorization, Congress asked us to take a look at that exact issue about whether good facilities could be inspected less frequently and asked us to do a demonstration project or program on that.

We did evaluate a number of facilities that had been basically significantly citation free and had them inspected every other year, and the results of that were placed on our Website, and what they tended to show is that even the good facilities in the second year of inspection were found to have an increased number of citations compared to even the standard facility.

So it has been looked at at least to some degree, and at least the preliminary results were not very conducive to the concept of having every other year inspections.

CHAIRPERSON HENDRICKS: Thank you very much.

Any other questions from the panel members, the speaker or the audience?

(No response.)

CHAIRPERSON HENDRICKS: In that case, I think we'll move to the break. We're scheduled on the agenda for a 30 minute break or for a 15 minute break. So we'll reconvene -- I just doubled the break -- so that we'll reconvene here in 15 minutes.

Thank you very much.

(Whereupon, the foregoing matter went off the record at 10:24 a.m. and went back on the record at 10:45 a.m.)

CHAIRPERSON HENDRICKS: To start the next discussion, I'd like to invite Dr. Helen Barr to the podium. She's the Director of the Division of Mammography Quality and Radiation Programs, and she's going to lead a discussion of the Institute of Medicine recommendations.

Dr. Barr.

DR. BARR: Thank you.

Good morning, everyone. First and foremost, I'd like to thank you all on behalf of the division as well as the office, and indeed, all of FDA for being here, taking time out of your busy lives and schedules to come and give us your thoughts and opinion.

And as you'll hear when I tell you a little bit about my background, I have been out in the real world. So I do know what it's like to come from there, and I can't tell you how much we appreciate you all being here.

First of all, before I start, I wanted to make just two minor corrections. Mr. Divine mentioned that we will be in the process of collecting dose data that the physicist supplies to compare to dose data that the inspector actually measures during the inspection.

Actually we are already doing that so that we can make a comparison because one issue we've heard is that there's possibly a disparity between the measurements a physicist makes and the measurements the inspector makes, and I wanted to indicate that we will be doing that, but we already are underway doing that, and we'll be able to compare those results as we along.

Second, Dr. Finder mentioned that the inspection demonstration program was in the last reauthorization of MQSA. That was actually in the first reauthorization of MQSA. There has been a reauthorization since then. I just want to make that minor correction.

In the interest of transparency and so that you know a little bit about me because obviously my background informs naturally the way I work here in the federal government. I graduated from George Washington University School of Medicine and did my internship and residency in diagnostic radiology there. I did a fellowship there and stayed on there on the faculty for two years before I moved to Kaiser Permanente here in the mid-Atlantic region where I was the Director of the breast imaging services, including interventional procedures, for nine mammography centers throughout the mid-Atlantic region that performed well over at that time 60,000 mammograms a year.

We have the second stereotactic unit in the Washington metropolitan area. So I have experience in that area since the very beginning of the modality.

I came here to FDA -- I was just counting on my fingers -- I just passed my sixth anniversary here at FDA, and I came on as a Deputy Director of the Division of Mammography Quality and Radiation Programs, and in I guess about a year and a half -- I don't know how long -- became the Director of the program.

So that's who I am. Dr. Finder asked me before we start on the subject at hand to just mention to you a little bit about what we're doing about what was one hurricane and now is two hurricanes in relation to facilities and personnel throughout the Gulf Coast who may have been affected by the hurricane. We have a significant amount of information on our Web site related to what facilities can do in natural disasters and, in particular, what facilities in the Gulf Coast can do.

Probably the biggest thing that we're doing is helping personnel who are moving to other states be able to get employment at other facilities. We here at FDA are looking at the last inspection that the personnel would have been involved in, and based on findings from that, providing personnel with letters so that they can document other initial and continuing requirements so that they can go other places and obtain employment right now.

These are folks that have had records destroyed in the wake of the hurricane. So Dr. Finder just asked me to mention briefly to you that we were hopefully doing good things.

I'm sorry. This microphone keeps -- if I lower it to where it needs to be it keeps tipping.

I also apologize in advance that I am a great seasonal allergy sufferer. So I've got my tissues and my sneezing and everything else up here. So I do apologize for that. Even though I am Dr. Finder's boss, my constant begging of him to schedule this meeting after the first frost doesn't seem to have gotten me anywhere.

(Laughter.)

DR. BARR: What our job -- you can imagine what the rest of my days are like.

What I'm going to lead us through here for the bulk of today and tomorrow is actually marching step by step through the Institute of Medicine recommendations. There's a lot of material here, and we want to get as much of your input as we can on these recommendations.

Some of them my guess is will require basically no discussion. Some of them may engender a fair amount of discussion, particularly when we get to the part on modification of MQSA regulations. What I'd like to do is perhaps not get stuck on the wording so much as the spirit of what is being recommended to be changed because I personally think that some of the new recommended wording is perhaps as confusing as some of the old wording. So rather than get bogged down on wording, you know, perhaps we can agree on the spirit, and then if something needs to be changed in regulation then, you know, experts on writing that can take our thoughts and put it down in the proper language.

Any questions before we begin about anything or shall we just dive right into it?

Okay. Here we go.

The background for the Institute of Medicine report is that over the last three years, and particularly the time of the last reauthorization, a lot of questions regarding the quality of imaging interpretation in mammography have been floating around, you know, through articles, through public opinion in Congress, and Congress struggled with a way to perhaps look at what the problems in image interpretation might be before putting anything specific in the law or taking anything out of the law and regulations.

The Institute of Medicine is a part of the National Academy of Sciences, and Congress commissioned a study from them in preparation for the last reauthorization -- excuse me -- at the time of the last reauthorization of MQSA in reparation for the next reauthorization of MQSA in hopes that the information from the IOM report could be used in the next reauthorization to improve particularly image quality interpretation.

Congress at that time also commissioned a GAO report on access to mammography and a couple of other issues, and although I know GAO is busy working on that report because we've been working on it actively with them, we do not have the results of that report yet, but luckily we do have the IOM results. So we're going to go ahead and get started with those.

The Congress' intent that, based on commissioning a study, for the IOM to look at a step to increase of interpretation, whether current regulation should be modified, the effects of recommendation on access to mammography, and identifying steps to insure safe and effective use of other screening or diagnostic tools.

The report is called improving breast imaging quality standards, and it was done specifically by the Committee on Improving Mammography Quality Standards of the National Cancer Policy Board at the Institute of Medicine.

There were four major areas of recommendation that the IOM came out with, and as I said, this is a very long, comprehensive report. So we've tried to take their four major areas of recommendation, and we'll be marching through them step by step.

Now, one was improve mammography interpretations.

Two, revise MQSA regulations, inspections and enforcement.

Insure adequate work force for breast cancer screening and diagnosis, and improve breast imaging quality beyond mammography.

I'm going to start with the first of those major categories, improving mammography interpretation. The recommendations within this category from the IOM were to revise and standardize medical outcomes audit, to facilitate voluntary advanced medical audit with feedback to establish specialized breast imaging centers of excellence; to study the effectiveness of continuing medical education, that should say. I know we had that slide changed, but somehow it's here wrong again. That should be continuing medical education.

Reader volume, double reading, and computer aided detection. So I'll go through the first of those recommendations to revise and standardize medical outcome audit.

This is just a lot of information about the different forms of positive predictive value, and if I can skip ahead here, I'll know which one we should concentrate on, which looks like PV-2.

The proportion of all women recommended for biopsy after mammography, Category 4 or 5, that are diagnosed with breast cancer. So particular note is that value.

And also different definitions for different forms of false positive.

IOM recommends that the medical audit should include the calculation of three core measures: the positive predictive value two, the proportion of women recommended for biopsy after mammography, Category 4-5, or are diagnosed with breast cancer; cancer detection rate per 1,000 women; and the abnormal interpretation rate, women whose mammogram interpretations lead to additional imaging or biopsy.

The rationale that they include in the report is that MQSA currently does not require calculation of specific performance statistics; that all of these three things together would be more useful than PPV-3. It's easier to calculate. PPV-3 is easier to calculate than PPV-1 or -- excuse me -- PPV-2 is easier to calculate than PPV-1.

That additional imaging assessment not included in the MQSA audit. I'm not sure what that means. Let me go back to that. I don't know. I can't speak to that specifically.

So I guess we'll discuss those first. So I'll go back to the slide that has the overall recommendations.

(Pause in proceedings.)

DR. BARR: Not only is Charlie Finder a great Associate Director, but he's the best administrative assistant I've ever had.

(Laughter.)

DR. BARR: Which I tell him all the time.

So here is the medical outcome audit, and I'm going to turn over to Dr. Hendricks and Dr. Finder if you have any discussion on this matter.

DR. FINDER: Yeah, it's Dr. Finder.

Basically as reported in the IOM summary, we do not require any specific statistics as part of the medical audit. We do require that the facility identify and track all positive mammograms, and we identify those read as suspicious or highly suggestive of malignancy, the fours and fives. They have to make a reasonable attempt to find out what happened to those patients and include that in their audit, but we do not tell the facility what specific statistics they need to do, whether they need to do any calculations at all, in effect.

And the IOM is recommending that these three measures be included as part of the regulation that all facilities must do, and we're interested in hearing from the committee on what they think about this approach.

CHAIRPERSON HENDRICKS: We've got a couple of radiologists on the panel. If we could solicit your opinions first on how these recommendations would impact your practice, for example.

DR. FERGUSON: Scott Ferguson from Arkansas.

I see no need for adding increased mathematical calculations. It would be a burden on the system and I don't think would add anything to the system to increase the number of calculations that you have to make.

Where is that information going? I mean, who's using that information? What good does it do I guess is my question.

CHAIRPERSON HENDRICKS: Thank you. I think we'll move along a little bit and maybe part of your question will be addressed as to what the IOM thinks about who should be using this information.

DR. FINDER: Any other comment?

CHAIRPERSON HENDRICKS: The next recommendation under this category is performance measures should be stratified by screening and diagnostic mammography. Rationale is difficult to interpret and compare performance with current literature or established databases.

Any comments? Discussion? I know we heard some in the ACR. Dr. Lee gave some opinions on this.

DR. MONTICCIOLO: This is Dr. Monticciolo. I'm a radiologist.

Yeah, I agree with the comments that were made earlier. I think it's very difficult to start discriminating between screening and diagnostic when there's differences among practices, what somebody considers a screen versus a diagnostic.

And so, like Dr. Ferguson next to me, I'm not sure how useful that discrimination will be and how that will help anybody. And I'll just add burden where I don't see much gain.

DR. BARR: Thank you.

MS. MOUNT: Carol Mount.

I agree. Within our facility we have a number of satellite facilities, and every one of those facilities also call screening and diagnostics different. So I think it would be very difficult to differentiate.

CHAIRPERSON HENDRICKS: I'd just add a comment. I think I'm a medical oncologist. So like the majority of my patients have breast cancer, have been diagnosed and treated for breast cancer.

I actually think that this may be a very important point. I think that just the fact that amongst the panel members out there is a great difference, and it has been the definition of a screening and diagnostic mammogram doesn't mean that we don't need to establish one. I think the facilities in this community -- and I practice in Bethesda, Maryland -- are really overburdened right now from women who are seeking out diagnostic imaging, and they just don't have the resources for the radiologist to read those films in prime time.

I think that if the public could be educated, the word well, the women that we are targeting for screening could be educated on what the true definition of a screening mammogram, which in my clinical practice is the screening of a woman age 40 or older with no breast symptoms at the time examination is done.

I think this is a very important public health issue, and I think it could really lessen the burden that certainly the facilities in this area are overwhelmed with women seeking out diagnostic imaging when really they are more appropriate for screening.

So I don't think that we should abandon this idea that we could level the playing field and create a definition of a screening patient that all facilities could accept. But I'd welcome other comments about that.

DR. MONTICCIOLO: Well, I think the problem comes in, just the variations of practice. I have one surgeon who wants all of his patients with cancer to be diagnostic, and so we fight this battle every year, and we have another surgeon when I was at Emory. The main surgeon, who was a well known, a nationally known cancer surgeon, said, "I want my cancer patients to be screened because I want them to feel as normal as possible."

Now, you have the implant patients. You know, should they be screening or should they be diagnostic? And I think you're right. It would be nice to have a standard, you know, who falls in where, but some facilities can't respond to those standards very easily. So we have to keep that in mind. We put more restrictions on facilities about what they can and can't do.

For example, implant patients. We now can't do them as screening. So we have women that have to drive 40 miles -- I live in central Texas -- to get their diagnostic mammogram and they have no breast complaints, but they happen to have implants.

So there's all of these variations, and I think you're right. If we had something more standard, but when we do impose that standard it's going to have implications. So I'm a little concerned about access and the difficulty of putting more layers on what we do.

CHAIRPERSON HENDRICKS: Yes, ma'am.

MS. PURA: Linda Pura.

We use simple terms as asymptomatic, symptomatic, and then special views for the implant. It could be as simple as that, and then, of course, you have to start subdividing, but those are simple terms that can be utilized.

CHAIRPERSON HENDRICKS: Also in response to the issue one of our tasks is to try to decrease the burden that inspection and mammography has placed on the system economically and clinically for the imagers. The concern about my patients who seek out or are continually in this diagnostic mode is the frequency with which they should be studied.

So a very high proportion of women are seeking mammography at intervals more frequently than years. Whereas if we could establish some standard or some period of time beyond which mammography more frequently than yearly could be performed in women who are long term survivors of breast cancer, for example, I really do think that the burden would be decreased significantly.

DR. BARR: Thank you.

DR. FINDER: Would they be screening or diagnostic?

(Laughter.)

CHAIRPERSON HENDRICKS: In my community here in Bethesda, the women are used to having -- what has become quite commonplace is women want face-to-face interaction with their radiologist. They'll call and they'll schedule because they know that a certain physician is going t be reading that day. Basically they want an appointment slot, you know, to meet their mammographer after their imaging.

And, of course, when you look at flow through a mammography unit, that can really cripple the flow and decrease the number of high quality images that the facility can read and the radiologist can interpret.

So, again, it's more of a public health issue to educate women and their families and their physicians on high quality breast care, you know, at the expert level because there are experts in images, of course, who understand the requirements and which women would benefit from more frequent imaging and which women require screening and diagnostic approaches.

DR. MONTICCIOLO: I agree with you. I'm in favor of increasing education. I'm just not sure that separating out audit data is going to help. To me that seems like an extra burden, but I like the idea of getting a more standardized approach to who gets screening, et cetera.

DR. BARR: Thank you.

C under recommendations option, that facilities should have the option of combining audit measures for physicians at multiple facilities. Their rationale in the report is that the data would be more meaningful or is more meaningful when larger numbers of exams per physician are analyzed.

Charlie, do you want to comment on what's currently the --

DR. FINDER: It's Dr. Finder.

I just want to kind of provide some background of where we are right now. Under the current regulations, each facility is required to do its own audit. So the data has to be done. Whatever data they do or whatever calculations they do, and again, we don't require any specific calculations, has to be broken down by facility and by individual physician at that facility.

Part of the reasoning behind that is we have authority over facilities, not over individual personnel, and that's the entity that we can hold responsible for making sure that that happens. Once you start expanding out to other facilities, it becomes more problematic.

Another issue was that since we did not require that the audit be done, either broken down by screening or diagnostic, we felt that if we could at least keep it to the facility level, then all of the physicians at that facility would basically be in most cases looking at the same populations, and they would be able to compare whatever analysis was done at that facility with the other physicians at that facility, and that was our purpose basically for the audit. It was not for a national collection or anything like that, but just so the physicians at the individual facility could compare themselves with the other physicians at that facility.

So in that sense, our regulations talk about these analyses being done facility specific. Now, we have adjusted to that. We actually have approved an alternative standard which allows mobile facilities in which their mobile units are each individually certified so that their own facility, but where the physicians are the same and these mobile units all go kind of round robin to the same populations.

We have allowed them to combine their data into one audit, but we have not done that yet for fixed facilities, and part of the reason is we couldn't be -- one, we didn't even get an alternative standard request for that specific issue, but the other is we do have concerns about how you're going to combine data from different facilities to make a cogent analysis.

If, for example, one facility is screening basically and another one is primarily diagnostic, what happens to the data when you combine those two?

But that's why you're here, and we want to hear what you have to say about this issue because it is being constantly brought up. As part of our guidance, we do suggest that even though according to our regulations you must base this on a facility, individual facility, we do recommend that practice groups that practice at multiple facilities combine their data and do a second analysis to get their data and look at that also because we do believe that the increased numbers can supply additional information.

But again, our current standard is the audit has to be facility based and then broken down by individual physician at that facility.

CHAIRPERSON HENDRICKS: Any comments?

DR. FINDER: Comments, thoughts? Do people think it would be a good idea if we allowed multiple facilities to combine their audits and just produce one set of data?

DR. MONTICCIOLO: It seems to me as a radiologist that what you're interested in is how the physician is performing. So if they read at multiple facilities, I see no problem with that. I would be strongly in favor of allowing them to combine data because the larger your numbers are, the larger the sampling and the more accurate look you're going to have at that person.

CHAIRPERSON HENDRICKS: I have a comment. In this area, in this geography, there is multiple satellite offices. So certainly I would support, you know, the data to be combined for multiple satellite offices when there's one large clinical practice responsible for providing the mammography services.

DR. BARR: Thank you.

DR. FERGUSON: My question would be are you talking about mandating or are you talking about allowing them to combine their data?

DR. FINDER: It's a very good question. It could be either one, depending on what kind of advice we get.

DR. BARR: And I think maybe D speaks a little bit to that. The recommendation is that audit data collection and analysis be verified at inspection, but not collected -- I assume they mean collected -- by the FDA, and the rationale is no change in procedure one because regulator is not able to verify the accuracy of the data.

DR. FERGUSON: I guess my question goes to this data is for the physicians to judge among themselves how good a job they're doing. It's not used for any other purpose, right?

DR. FINDER: Well, correct. Under the current regulations, the information obtained from that audit is supposed to remain at the facility. We do not collect that data. We do not create a national database or use that data except to see that it has been done. That's all we do.

DR. FERGUSON: And so I would favor allowing rather than mandating because this is for physicians to improve themselves and see where they're shortcoming, and I think they should be measuring those standards against one another, and if someone needs additional training or whatever, they take care of it.

But as far as mandating it, it doesn't go any further than the group. I don't see where you gain anything from doing that.

DR. FINDER: One issue that I would bring up is what would happen if some of the physicians are from multiple facilities and some of the other physicians don't have data from some of them. It becomes an inspection type issue that we have to at least look at to figure out the complexity of that.

Who do we end up citing if some data isn't there?

DR. FERGUSON: Well, and the problem I see and the reason I don't think it should be mandated is that one facility may have a whole different group of patients that may be doing primarily diagnostic, and I'm reading, and the other is doing screening, and another guy is reading. You combine that data, and you have a different subset of patients. That's why I don't think it ought to be mandatory.

DR. BARR: And as best as I understand the recommendation here, at least at this level of audit, and you'll see as we go along the recommendations are for different levels of audit, I interpret this to say, based on D, that we should allow facilities to do this, but there would be no difference in the inspection procedure or any citation for facilities that don't do this.

E in Recommendation No. 1 is increase reimbursement rates to cover new audit procedures. Rationale is costs are already significant. The new audit procedures will add to expense. Costs were not factored in past reimbursements, and health care payers should cover costs.

That probably doesn't require much discussion.

DR. FERGUSON: I definitely support that.

(Laughter.)

DR. BARR: Recommendation 2, and here you see --

CHAIRPERSON HENDRICKS: I'm sorry to interrupt. I just wanted to return before you move into the second set of recommendations because we have so many talented diagnostic radiologists here on the panel and also Dr. Barr.

If we're not able to accept these three metrics, for example, to try to establish some quality parameters in mammography, for you in the trenches who actually do this, which parameters are useful for determining quality of care, either a simple or more complex? What can we offer in lieu of?

If we do not accept these recommendations for the reasons that were stated in the discussion we've had so far, what is a good surrogate if one exists?

DR. BARR: I do think your question is an excellent one. Id o think it might be helpful if I run through the next levels of audits and bring out different parameters, and then perhaps we can discuss this as a whole on point to your question, which is very well put.

Although this is a different recommendation, it still relates to audits, and this is a voluntary advanced medical audit with feedback. So we sort of see the baseline that we talked about and then this, and then we can address Dr. Hendricks' question.

In this recommendation of a voluntary advanced medical audit with feedback, the recommendations that the audit should include collection of patient characteristics and tumor staging from pathology reports. The rationale is to record more useful data from pathology reports, such as tumor size and lymph node status, record patient characteristics, such as age, family history, breast density, presence of prior films and time since last mammogram.

I'm just going to run all through this and then we'll go back.

Establish a data and statistical coordinating center to electronically collect, analyze and report advanced level audit data and provide regular feedback to interpreting physicians.

I think we see here why there were all the cost recommendations. I don't think it pertains so much to that initial recommendation, but to some of these more advanced ones.

Develop, implement, and evaluate self-improvement plans for interpreting physicians who do not achieve benchmark performance, and aggregate summary data on interpretive performance, including recall rates, PPV-2, and cancer detection.

Under the same recommendation would be to test different methods of delivering audit results to interpretive performance, study randomly selected facilities using required basic audit procedures for impact on interpretive quality, protect quality assurance data from discoverability, and the rationale for all of this. The statistics and analysis group needed for uniform feedback to improve quality, studies needed on feedback to improve performance, national benchmarks needed for facilities to assess performance. It would test the impact of basic audit procedures, and the Breast Cancer Surveillance Consortium and the Agency for Health Care Research and Quality should be utilized as they are viable models for data collection procedures.

And, again, the report stresses several times the discoverability issue.

So now back to Dr. Hendricks' question, I think we could use some input on, you know, what you think of the ideas in these recommendations, and if you don't think that these are the things that necessarily need to be looked at to improve performance are the things that you do think are necessary to be looked at and collected to improve performance.

CHAIRPERSON HENDRICKS: We'd also like to invite comment from the patient advocacy representatives on the panel, from their perspective.

MS. PURA: It's interesting. We're going through this right now with our primary care clinicians in attempting to get them to report tumor size and axillary lymph node status, et cetera, and this is required by the CDC for payment for our state program.

Just getting the reports are unbelievable. I mean there are various routes one can go, but surgeons notoriously do not return this information to primary care clinicians. I can't see how they'll even return it to a radiology group, more or less our primary care clinicians who may have some input into treatment.

I don't know where this information would be very valuable in a radiology audit to see capabilities of the radiologists themselves in practice. Hopefully they'll be identifying tumors. I don't know if they need to get into axillary node and if they need to get into staging and tumor size, et cetera. I don't see where that, in fact, has anything to do with their quality of practice, and obtaining that information may be very timely and very costly for them.

I'd like to see how everybody else feels about that.

CHAIRPERSON HENDRICKS: Yes, Dr. Lee. Could you step forward to the microphone, please, and reintroduce yourself to the group?

Thank you.

DR. LEE: I'm Dr. Carol Lee. I'm from Yale University, and I also represent the American College of Radiology.

I respect your comments, but I disagree. I think it's very important for radiologists to know the stage of the cancers that we detect because we're only picking up large cancers that have already spread. We're not doing a whole lot of good, and the goal for mammography and one of the indications of quality is that we do detect small, treatable cancers. So that is very important information. I agree completely with the difficulty associated with the collection of that data and also, if I may just make a comment about what I believe is some useful metrics, what we want to know is the cancer detection rate, and we want to know what our false negatives are, and right now the discoverability of the false negatives is very difficult.

There are no well established, widespread mammography registries. Tumor registries exist, but they are hard to access that information, and I think these are all issues that can be hopefully addressed.

MS. PURA: Again, I agree with Dr. Lee in evaluating size and so on, and it's very important to know, but finding that information has become a real difficult problem, and I am very concerned. The information is absolutely important to the diagnosis, and the ability of the radiologist, but I am concerned about them getting that information, and will that, again, cause the access to radiology to go down because of another stringent regulation on them.

DR. BARR: I think one thing the Institute of Medicine is trying to get to here is that the way the audit is now it's fairly basic. You know, you need to do an audit, and we go in at inspection time, and ask you if you've done the audit, and that's about it.

Do you think that in regulation there should be more requirements for what the FDA looks at as far as what's been done for the audit? And should that information just stay at the facility? You know, what should be done with it other than the inspector seeing that it has been done? Do you think there should be citations for people who don't do these things?

I think that's what they're trying to get at, is the audit is pretty basic, right now what FDA requires, and is there anything else that you think is vitally useful in that arena.

CHAIRPERSON HENDRICKS: That's a good background. I think the way to think of this is what's happening across health care. Certainly hospital based health care in the United States is the creation of these report cards for hospitals. They don't like them, but they're certainly out there and maybe even before their time, but I think we should think about we know -- we can recognize good quality mammography facilities and poor ones or maybe ones which don't have as high quality. So we have to think about what would be on a report card and how they can be evaluated without, you know, basically shutting them down by creating some onerous regulations.

But yet we know that we need to be able to evaluate them and compare them to one another. So what would be on a mammography report card to identify an A-plus facility compared to a facility that is marginal or offering poor quality imaging?

Dr. Monticciolo.

DR. MONTICCIOLO: I guess I'll just make a comment. I realize that the auditing right now is under a lot of scrutiny. It's hard for me as a practicing mammographer to think that adding these additional burdens is going to improve interpretation because I don't see any evidence that ultimate patient outcomes are affected.

I think what we're doing now is we're trying to look at certain benchmarks in our practices to see how we compared to each other in a certain practice setting, and those are somewhat useful, but the extent of auditing here is going to be tremendously cumbersome, and I'm not so certain that it's going to affect the interpretative ability of the physicians involved.

And so I'm not convinced that there's evidence of that, and that's why I'm not a big fan of adding more and more layers to how much we collect and look at. There's only so many hours in the day and already my colleagues who are not trained in breast imaging, they usually follow my lead, and so whatever mistakes I'm making, I guess, are being multiplied, but you know, usually I'll set the standard, and I've pushed the standard up for people who are just kind of doing other things and doing a little bit of mammography, but you know, I look at this and say, "Well, if we add more and more layers, I think more and more people will just drop out."

And the people who want to do a good job, I think, are personally driven to do that, not that they shouldn't be looked at, but over scrutinizing them and collecting more data I don't think is going to change their interpretation tremendously.

And I agree with Linda's comments about tumor staging. These things are important, but to acquire that data and to really try to dig this up and put out reports every year on it is further going to diminish the desire of people to enter the field.

So I'm very concerned about that. So unless something is really proven to improve the interpretive ability of physicians, I'm not in favor of just laying it on and hope that it would help.

CHAIRPERSON HENDRICKS: Yes, Ms. Holland.

MS. HOLLAND: Jackie Holland from Ohio.

I'd like to know if anyone can tell me what the rationale was in the first place for the FDA to verify that it had been done but not collected. I don't understand why they were even verifying it if nothing was going to happen and if it didn't really affect the inspection.

DR. FINDER: It's Dr. Finder.

The rationale behind it was that this information which hadn't been required of facilities in the past was to be used by the facility. In fact, we do have regulations that talk about the audit interpreting physician who oversees this process, and part of their responsibility is to get back to the individual physicians involved in this audit with their results and talk it over with them.

But we did not in regulation specify what actions were to be taken or what were the benchmarks or what analysis was to be done. It was supposed to be an educational activity for the facility to improve on their own without getting into the specifics and telling them how and what they had to do.

And the recommendation from IOM is to get a little bit more specific in terms of what they should be doing. My understanding at least on the simple audit, the general audit would be that there wouldn't be much other change. We wouldn't collect this data for a national database. It still would remain within the facility, but it would be more standardized for them within that facility to look at their own data.

I will tell you some of the arguments that were brought up at the original time when we were talking about audits and why we didn't ask for specifics at that time were that any statistical data that you might obtain is highly dependent on a number of factors, including volume, and one of the worries that was brought up at that point was that you might have a low volume reader whose numbers could be bouncing around all over the place, and it wouldn't mean that there's any real change. It's just the statistical variation that occurred.

Another was the business and discussion of what constituted screening and diagnostic because the baseline benchmarks for those two groups are different. Populations are different. You know, if you're dealing with one population group versus another, the incidence of cancer can be significantly different in those, and trying to compare over the country certainly would cause problems, although again by trying to limit it to just a single facility, we were hoping to kind of minimize that variation because everybody reading at that facility presumably would be looking at the same population.

So that's the history.

DR. BARR: And, Ms. Holland, this is Dr. Barr.

I don't really think that if you don't do the audit you don't get cited. If there's no evidence that you can provide to the inspector at all that an audit takes place, then you can get a citation for that, but there are no specific elements other than that it has to be divided up by physician, and we're asking if you all think there are any other specific elements that should be in the audit that the inspector would specifically take a look at to see if its' there.

I don't want you to think you can just totally not blow it off and not have some consequence. You can get a citation, but other than that, there are not a lot of specifics in it, and I think that's what the IOM is trying to get to.

DR. FERGUSON: I'd like to say I agree with the other two panel members that I think this would be a burden that we don't need, that will not help the interpretive skills of the physician, which looking back, that's what they wanted to know. How can we help with mammography interpretations?

And I'll say the audit -- I'm thinking back when I first started doing my audit is when it was required, and it has helped me personally to look at my numbers and to hopefully improve every year and see what I missed and go back and see what I missed and why I missed it, and it has helped me improve in my interpretation.

So I think that the audit that we have -- and I was surprised, like she says. Why did we have an audit and it didn't go anywhere? It has helped me personally as an interpreting physician. And should it go any further? I don't know.

MS. PURA: Have there been any benchmarks that have been offered by either ACR or by BIO of IOM for this particular categories in various staging? Has anything come out, Dr. Finder, that you know of?

DR. FINDER: In terms of benchmarks, there have been a number of publications that talk about various benchmarks for screening for diagnostic and for mixed facilities. Dr. Sickles has done an article and talked about various benchmarks.

Even before MQSA, there was a publication by the AHCPR, which is now AHRQ; a study done by the federal government, a nongovernmental agency of the federal government, whatever that means, published some benchmark guidelines that can be used by facilities.

So, yes, there is information out there where you can kind of compare yourself against some kind of national standard, but it doesn't really take into account the variation that can occur in an individual facility, and if a facility wants to look at that data and compare itself to it, it's truly on an educational basis, whereas if it was mandated that there be some benchmark, then that's a whole different story.

But there are numbers that facilities can look at.

DR. BARR: Okay. So I think in summary what I'm hearing then is that we should continue at inspection to look that an audit has been done, that it should be by individual physician, but that we should allow facilities who can and want to to combine audit data across centers to look at larger numbers.

That's what I'm hearing so far, and we'll go on to the -- there's a little more in this advanced audit piece.

Recommendation 3 is to designate specialized breast imaging centers of excellence. The first part under that in the report is that these centers will participate in basic and advanced medical audits and test approaches to improve quality and effectiveness. They would test effects of high volume, double reading, quality assurance, patient reminders. They would develop and evaluate interpretive skills assessment exams.

The rationale behind these recommendations was stated that several countries have integrated centralized breast cancer screening programs, but in the U.S. screening is decentralized and offered in diverse practice settings.

These excellence centers could provide multi-disciplinary training and work environments for diagnosis, could increase job satisfaction, retention of practitioners' productivity, and quality of the breast care team.

High quality facilities could attract high quality personnel. Incentives for becoming one of these centers of excellence would be similar to what was stated previously: high reimbursement rates, and could be used to recruit patients and referrals. I guess you would be allowed to put out that you're one of these centers of excellence.

Rationale that supportive elements and incentives are critical to encouraging facilities and personnel to strive for higher quality. These centers should serve as training centers for breast imaging and regional mammogram readers. The centers would have the expertise to develop and host training programs in imaging.

Interpretation at centralized facilities could help alleviate access in low volume areas. The centers should be linked with facilities that provide comprehensive and multi-disciplinary breast care. The rationale is that imaging based centers need continuity with facilities providing non-imaging breast care treatment and follow-up.

And so any comments on these breast imaging centers of excellence?

CHAIRPERSON HENDRICKS: I'll start out with a comment. I just wonder, Dr. Barr, in your opinion and with your familiarity of mammography facilities in the United States, which centers do you think are already meeting these criteria, if any, or how many?

DR. BARR: Well, I think that gets back to like a score card like you said or a report card of facilities, and you know, various states have tried to market facilities as being, you know, in the upper echelon or in different strata and, you know, have found huge problems with doing that.

The best that I can tell you from our data is that 70 percent of the mammography facilities in the country practice quality mammography as defined by MQSA, and you know, that's the best I can tell you right now.

DR. MARTIN: Dr. Hendricks.

CHAIRPERSON HENDRICKS: Yes.

DR. MARTIN: Melissa Martin.

As a consulting physicist, I see what I've usually referred to as the good, the really good, and then the ones that barely meet the criteria, and I would just highly encourage us to or encourage the FDA to pursue this idea because there is definitely a vast difference in the quality of care out there, and I think we do need to encourage this development and designation for those centers that are doing upper level quality care.

And if you ask me, I would say, well, we currently cover around 300 facilities, and I would say probably 100 of them definitely meet it already, but they are definitely doing more, personnel-wise, skill-wise, education-wise, than the local stand alone unit that does screening only, and I think we need to differentiate what those facilities are doing.

DR. BARR: And, Ms. Martin, would you put that information out publicly and for patients that didn't have access to such a facility, what would you tell them?

DR. MARTIN: To encourage them to get access to that level facility.

DR. BARR: Even though such a facility that you're describing might not be available to them.

DR. MARTIN: Well, I function in a very crowded area, and I find it very frustrating sometimes that we have the equivalent of a center for excellence, and three blocks down the road we have a minimally qualified facility that is still in practice and getting paid the same as the facility that's a center for excellence.

DR. BARR: And at one time we say under MQSA, you know, we're talking about standards across the board so that any facility who meets them, you know, meets the criteria.

CHAIRPERSON HENDRICKS: Another question might be if we propose to the mammography centers of the United States whether they wanted that designation, how many would voluntarily want to undergo the steps that it would take. What is your feeling on that?

Because we've heard that the current basic, you know, bare minimum audit is burdensome. So do you see that there would ever be any desire for even the excellence centers to get this designation?

DR. BARR: You know, I think a lot of it depends on what a lot of people have already said in how much money would be available to centers, how the reimbursement would be affected by, you know, if you could receive higher reimbursement for doing this, if Congress is going to give money for doing this.

But like a lot of things, I think people might be loath to do these requirements because they don't have the money or the manpower to do it.

I also worry about, you know, the woman in rural North Dakota who doesn't have access to what people -- I don't even think we have the criteria for what one of these centers of excellence is, but I also worry what we tell the people who, you know, don't have -- if facilities become these centers of excellence, how do you get access to them?

You know, perhaps as the digital age gets more advanced, that problem might be decreased, but right now I do worry about what we would tell patients who would say, "Well, does that mean the center I go to isn't good enough?"

You know, MQSA, there's a certificate on the wall. My center has had no violations that I'm aware of. You know, does that mean I'm not getting good care, that I've got to get on a plane and fly somewhere to go to one of these centers of excellence?

CHAIRPERSON HENDRICKS: Yes.

DR. WILLIAMS: This is Dr. Williams.

With respect to the question of being able to afford establishing centers of excellence, I know that many academic institutions have lots of centers of excellence, cardiac centers of excellence, digestive centers of excellence, and many of these programs have been to a certain degree underwritten by grants from the NIH.

And one of the things that would be worth considering is whether some of the funding, whether it's NCI or someone else, would be interested in putting out specifically RFAs for establishing these centers and with perhaps the express statement that there would be funding written into the budgets for assisting access to these centers for women who are not located necessarily right next to them.

DR. BARR: Yeah, I think that's an excellent comment. Thank you.

One thing I'd just like to point out is as far as I know most of those other kind of centers of excellence, you're not talking about a screening modality, and I think that that, you know, plays a role here.

Yes, Dr. Ferguson.

DR. FERGUSON: I agree that a designation of a center of excellence will cause burdens in more rural areas like mine. Women will say, "Well, I have to seek this facility," and facilities who are doing good quality work will dry up and you will lose access.

Like Dr. Williams says, I think incentives are an excellent idea for people to try to attain this, and the incentives in the form of grants or increased reimbursement for facilities who meet these criteria are excellent ideas, but to go out and designate them, still continue to pay everybody the same and say one is better because they provide training and multi-specialty facilities, I think, would ultimately harm access to quality care that is out there.

CHAIRPERSON HENDRICKS: If I could interject before we go to Linda who had a comment, a lot of the members of this panel don't have to deal with payers, but the way that the big payers in this community are headed, I think as this whole idea of pay for performance.

I think every big insurance carrier in the United States is very much interested in reimbursement and lowering reimbursements for some services, but increasing reimbursements for what they're certain is high quality medical performance. So that's a little bit of a circular argument because if we go to the payers to ask for support and increase reimbursement for a breast center of excellence or even a center that has met all of the criteria for our audit, we have to go to them with some metric to demonstrate that, in fact, we should be paid for excellent performance.

Linda, you had a comment?

MS. PURA: There's pros and cons, of course, for the centers of excellence, but in the milieu that I live in and work in, the women that we see, if we can have and be so blunt to say a one shop stop that has many, many procedures that are offered and women don't have to come back, that eases some of the access to going to various and sundry places to get the procedures that they need.

I would, of course, want to see that a center of excellence does take the Medicaid patients. That's another major problem that we are finding now, is that centers are refusing to take, as I say, our women, and so that would be, if I was looking at a facility, that would be something that I would want to see.

However, I don't know if we have any impact on the federal reimbursement for Medicaid at all.

CHAIRPERSON HENDRICKS: We have time maybe for one more brief comment on this topic before you break for lunch.

DR. BARR: That sounds good. I'd like to point out here that I think one thing that IOM is saying in these centers of excellence is, you know, not only would it be a designation that patients could use, but that these centers would be the ones that would test out the different things that are now on the table that might improve quality: the high volume, the double reading, different things like that; that these centers would sort of be our researchers, as it were, into what things might improve quality.

CHAIRPERSON HENDRICKS: Yes, from the audience, the final comment before lunch. Please introduce yourself.

DR. SHOPE: Yes. I'm Tom Shope. I'm with the CDRH.

I'm not directly involved in the mammography program in great detail, but it seems to me like it's worthwhile making one comment here, and that is the discussion of this Institute of Medicine report is the report was made to Congress. It was a report about the national mammography situation and what Congress ought to do in order to improve mammography, and so it's not a directive to FDA to do all of these things.

And so I just wanted to say when it talked about a voluntary additional medical audit kind of thing, the first word there was voluntary. I mean, that seems to have gotten lost in the conversation here, that they were suggesting there be some mechanism set up to perhaps provide some recommendations as to what a good quality audit might look like in a facility and some way to encourage facilities to implement these things.

I don't think there was any requirement that FDA make this mandatory, and the same thing here with the imaging centers of excellence. It sounds to me like a recommendation to Congress from the IOM that Congress consider how could we foster the establishment of these kinds of -- and I see them as research facilities -- to look at the effects of the various things that one might do to improve mammography, not necessarily that FDA would require all of the facilities to do these things or that we would set up criteria for when you qualify to be one.

I think what Congress was doing is saying we need to have some ways to encourage the establishment of these things. The IOM was saying to Congress that which might, of course, get into the issue of who would fund them, how would they be established, all the research activities that need to go on.

So I don't think it was a message that FDA necessarily needed to do. Pardon my butting in, but I think it seemed like there was something he had missed here.

CHAIRPERSON HENDRICKS: I appreciate that comment.

And with that, I think we'll take a break, and then, of course, we'll be resuming this same discussion and working our way through the document after a one hour lunch break.

We'll return then in one hour and 15 minutes. We'll reconvene at one o'clock.

(Whereupon, at 11:47 a.m., the meeting was recessed for lunch, to reconvene at 1:00 p.m., the same day.)

AFTERNOON SESSION

(1:04 p.m.)

CHAIRPERSON HENDRICKS: I want to call to order the afternoon session.

We're going to resume the discussion that we held this morning with Dr. Helen Barr helping guide us through a discussion of the Institute of Medicine recommendation beginning with Recommendation No. 4.

DR. BARR: Thank you and welcome back.

Before I go on to Recommendation No. 4, I just wanted to make a very brief comment about Dr. Shope's comment from the audience which we ended with when we broke, and you know, he's perfectly correct. This is a recommendation to Congress. I was actually going to talk about that a little bit later when it becomes abundantly clear that it's not FDA; that it would take, you know, a multitude of HHS and other agencies and other venues to institute some of these things if they were to be.

However, with that being said, Congress will definitely be looking to FDA, especially on the regulation part. You k now, I think the biggest danger we all feel is that Congress will expect these things to be done without appropriate monies, incentives, et cetera, along with it.

Recommendation No. 4 under the section we've been working on is to study the effectiveness of continuing medical education, reader volume, double reading, and computer aided detection.

First, the recommendation is to demonstrate the value of CME for improving interpretive skills. The report cites the rationale as this would enable interpreting physicians to identify weaknesses and take steps to improve interpretive performance. We could continue to develop innovative teaching interventions to improve interpretive skills.

Anybody want to make any comment on demonstrating the value of CME for improving interpretive skills?

I think this is particularly important because at the time of the last reauthorization we almost had in the reauthorization, but didn't get a proposal that was on the table to make five of the 15 CMEs for physicians that we currently require into self-assessment type CMEs.

And I think it didn't go on the table because people raised the question that we didn't really know the value of CME in improving mammography interpretation. So I think this is an important area for comment.

DR. MONTICCIOLO: Can I make a comment?

I just wanted to point out that I think it is an important area, and I don't know that we'll need to address it because the American College of Radiology and the American Board of Radiology are heading toward the maintenance of certification to allow people to keep their licenses, and part of that will be a requirement to have self-assessment modules. So we'll have to have two every year over the ten years of practice.

So already that's going to be mandated to keep your radiology license. I think that will be taken care of with that.

It's not directed specifically at mammography.

DR. BARR: Do you mean for your board certification?

DR. MONTICCIOLO: That's correct.

DR. BARR: Yeah.

DR. MONTICCIOLO: And so I think even if people that have unlimited certificates will probably end up adhering to that program just because of reimbursements, et cetera.

DR. FINDER: This is Dr. Finder.

I want to bring up that point later because we actually have a discussion point in our guidance that we're going to try and discuss this issue about expiring board certificates. So that is an important issue that we will hopefully not forget about later.

DR. BARR: Thank you.

The next recommendation is to determine the effects of reader volume on interpretive accuracy, the rationale being currently there's insufficient evidence to recommend an increase in minimum interpretive volume. No basis for specifying a higher level of reader volume, and again, I think an important area to comment on, particularly when our charge is to put things into effect that wouldn't affect access.

So we appreciate your input.

DR. FERGUSON: I would agree that I think the number is sufficient at this time in order to insure access. There are physicians who don't read as many as others, that do a very good job, and you know, 400 and whatever it is a year I think is sufficient.

DR. BARR: Recommendation C is to look at the impact of double reading in CAD on interpretive performance over time in different practice settings and at different levels of experience.

Rationale here is cited as a second look by another reader or computer program not verified by prospective clinical trials, and effects on specificity are not fully understood.

CAD programs are being refined. So effective use could change over time. Studies use -- and I guess give us studies needed on effectiveness findings could help us use the information more effectively. Studies need to confirm, if consensus, double reading may be most effective.

Here we go again.

CHAIRPERSON HENDRICKS: Comments from the audience?

DR. BARR: In other words, I think that IOM is suggesting that there's important things that may go into interpretation, but we don't have enough information yet.

CHAIRPERSON HENDRICKS: We have a comment from an audience member. Please identify yourself.

MS. WILCOX: Pam Wilcox, ACR.

These recommendations for studies seem very important to impacting ongoing quality and knowing what tools we need, but there doesn't seem to be any way of addressing the funding for these studies or where they're going to come from in the IOM report.

Has FDA had an opportunity to think about that or look for opportunities for funding for any of this, or is that what you're seeking from your committee?

DR. BARR: As we mentioned before, you know, these are recommendations to Congress, and hopefully Congress will be addressing where funding for these types of things would come from.

MS. WILCOX: So are you sort of seeking input from this community to point which ones you really want to push to Congress to get funded for studies?

DR. BARR: Yeah, I think we're seeking input of which of these things, you know, do we think might affect interpretive skills, if any, and are they worth studying. Do we have enough information now on any of them to require them? You know, do we need to study them?

I think the funding questions are obviously right up there on everybody's mind.

Thanks, Pam.

Anybody have any comments on the funding issues or if any of these are worth studying? I mean, it certainly seems that before we get something in regulation, it's my understanding from people that we would like to have data that shows that any regulations that we get are worthwhile having and are on point to the task at hand of improving interpretative skills.

MS. MOUNT: Carol Mount.

From what we have evaluated at our institution, I think that the CAD program would be something that would be definitely worth pursuing. We have run our own study and found that it did increase the early detection rate by having the CAD.

DR. BARR: Thank you. That's good information to know.

If anybody else has experience with CAD that they'd like to share.

Charlie, do you have any?

DR. FINDER: Dr. Finder.

I just wanted to point out a couple of things in terms of the past history. It's interesting to note that as Dr. Barr mentioned earlier, some of these items that IOM looked at were issues that were brought before earlier versions of this committee in terms of possibly implementing these as regulation, the idea of making some of the CME and interpretive skills type CME. Raising the number of mammograms read over a period of time has certainly come up many times. The issue of double reading has been discussed many times, and certainly CAD is one of those, and it's just interesting that the IOM when looking upon this didn't feel that there was enough evidence at this point to actually make any recommendations to FDA to actually implement any of these things.

My question to the people here is: what type of evidence do you think would be useful for somebody in the future to decide whether these were actually useful things to implement or not? Does anybody have any idea of what type of research, what type of study could be done, not necessarily that FDA would do it? As has been pointed out, this is an issue that Congress is going to hopefully eventually decide will be looked at by somebody, but not necessarily FDA.

DR. WILLIAMS: This is Dr. Williams.

On the topic of CAD, I think probably quite a number of groups would agree that that's something that kind of needs to be looked into and is being looked into, and it shows a lot of promise.

I think there has on the topic of funding, there have been a number through the years of very well funded basic studies on the effect of CAD, and what it sounds like we're talking about now is a fairly large multi-center trial that would evaluate the effectiveness of CAD across a variety of different types of institutions, and the thing that springs to my mind there as one possibility would be Akron.

Akron, as you know, one of its charters is to evaluate the early efficacy of diagnostic tools, and now having the DMIST trial just wrapped up or not wrapped up, but the first results now out, that might be a reasonable thing to think of for the future.

DR. BARR: Thank you.

I think those are on-point comments, and I think these things are also going to lead into people now, for example, CDC and its breast and cervical cancer program, you know, paying for CAD or continuing to pay for CAD or increasing paying for CAD, you know, paying for digital mammography in their program.

So a lot of these things are, I think, important issues. Okay. So I guess what I'm basically hearing on this part is that we don't have anything specific. No one is coming forward and saying, yes, X, Y and Z are the things that we know improve a radiologist's interpretation, but rather there are a number of areas such as reader volume and computer aided detection that we need to continue to study.

I think the Chair can recognize a speaker from the audience. Before we get too far from audit, I had someone approach me who might be able to shed some more light on audits and mammography situations. So if the Chair might recognize, welcome and introduce yourself.

MS. MYERS: Hi. My name is Susanne Myers, and I'm the Senior Vice President of Mammologics. We have been in business for about ten years now, and we assist mammography facilities with auditing, patient tracking, the notification letters, and the reminder letters, and I really just wanted to mention one of the things that we do in the auditing process, and this goes back to the discussion about screening versus diagnostic, which is really asymptomatic versus symptomatic, is we feel it's very important to make that distinction because in order for you to really understand your practice and really understand the audit data, you have to know the mix of your patients, and that will assist you down the line, and when you're looking at your data to kind of get an idea of what the numbers mean.

We currently have about four million breast imaging procedures in our database and we've been assisting mammography facilities with compliance issues. One of the things that I just want to point out is that a lot of mammography facilities really want to do a better job, but I think there's a lack of information out there for them to do a better job, and I think from a guidance standpoint, I think we could really help with coming out with some guidelines as far as the auditing requirements. I think we could really make a difference in the quality of the mammography services that we are rendering at this time.

And that's really all I wanted to say.

DR. BARR: I thought one thing that you said to me was interesting, that you have facilities that do 1,800 mammograms. You have facilities that do 18,000 mammograms, but yet you can, you feel, provide them with significant audit data that can be used in ways.

MS. MYERS: One of the things that we kind of help our clients with is to look at their data over time, and a lot of our clients that really are interested in doing a better job, they have implemented quality improvement programs when they actually use the data on an ongoing basis to monitor what's going on in their practice.

So when you're looking at desirable goals, and that came up before as well, is there desirable goals that facilities should be striving for, and obviously it depends. It depends on your patient mix. It depends on how many radiologists you have, how many facilities that you're servicing. So all of these things need to be taken into consideration.

DR. BARR: Thank you. Appreciate the comments.

MS. MYERS: You're welcome.

DR. MARTIN: Dr. Barr, I know several of us -- I get involved with several facilities when they say "help with the audit," and I think one of the frustrations on the facilities end of things has been as you have said. At this point the audit has been very wide open. As long as they, quote, performed an audit, it wasn't really specified what was in it, and I do think that would be a recommendation if some form were at least a minimum of what information was required for the facilities to have in their audit, but again, I think we need to figure out how we're going to specify, if at all, what we do with the data when we get it, and that's what the radiologists on the panel, I think, have been saying.

Benchmarks are going to be totally dependent obviously on the patient population that you work with, and at least the feedback I'm getting from most of the facilities is they're very antsy about having that number, the magic number set unless we have very clear standards and databases to go with, and frankly, obviously, that's not the physicist role. So we're not a lot of input just because of the math, and sometimes the facilities want help with it.

(Laughter.)

DR. MARTIN: So I'm looking for input, and anything that you can help us establish those standards or as the committee establishes those standards, it would be welcome information, I think, to most of the facilities if someone decides what is a minimum set of criteria that they have to have ready for an inspector.

DR. BARR: Thank you, and certainly, you know, we brought that up for discussion, and I directly asked, you know, what should be in the audit, what should the inspectors look at, and the only thing that I heard that there seemed to be a general consensus on was to allow combining of facilities' information to look at larger aggregate data.

Thank you.

Okay. Now a huge section that we're going to be dealing with and it's, again, a lot of information, and this one is going to be particularly tricky because we're going to be dealing with recommendations for added wording to the regulations, deleted wording from the regulations. So see if we can work our way through this.

The next section of the four big recommendations that I talked about at the beginning that IOM made falls into the section of revising MQSA regulations, inspection procedures, and enforcement.

Our Recommendations 5 and 6 that fall under this category is to modify the regulations to clarify intent and to address current technology; to modify inspections by streamlining processes, reducing redundancy, and addressing current technology; and to strengthen enforcement for patient protection.

So we'll start with the Recommendation No. 5, modify regulations to clarify intent and address the current technology.

What IOM recommends in general are the following, and then we'll be marching through specific regulations. I want us to remove the exemption for stereotactic breast biopsy procedures and develop regulations, and I believe -- correct me if I'm wrong, Dr. Finder -- that tomorrow we're going to have a more specific and dedicated conversation on this. So probably just generally skip over this particular point today because we have some speakers also on this issue for tomorrow.

To develop regulations for digital mammography; to update assessment categories to reflect BI-RADS, including the known biopsy proven malignancy; to establish luminance standards for viewing mammograms; to eliminate modality specific CME.

As we march through this specific regulatory text, IOM's recommendation to added text to the regulations will be in these kind of parentheses and green print, and their recommendations to delete text from the regulations will be the parens that look like greater and less than, and in the kind of peachy-orangy print.

As I said, we'll skip over the stereotactic discussion until tomorrow.

Develop regulations for digital mammography. Should develop a uniform set of quality control tests and test criteria. This should not preclude performance of additional tests recommended by the equipment manufacturer, and to update assessment categories to reflect BI-RADS.

So this is our Section 900.12 in the regs. The overall final assessment of findings classified in one of the following categories, and you see the --

DR. FINDER: Dr. Barr.

DR. BARR: Yes.

DR. FINDER: Can we go back?

DR. BARR: Sure.

DR. FINDER: On the develop regulations for digital and just start on that one.

DR. BARR: Oh, I'm sorry. Yeah. I didn't realize we didn't have anything specific read comments on that because we don't have any.

Yeah, Charlie. Could you go into where we stand now on the --

DR. FINDER: Right. Let me try and give a little bit of the history behind the current regulations, how we got here.

Basically, the regulations were developed before full field digital mammography, any of the units were actually approved for commercial use. So in order to address something that we really didn't know what was going to happen, we put in a regulation that said that when these new mammographic modalities come into existence, facilities would be required to follow the manufacturer's recommended quality control, and that's where it stood in 1997 through '99 when the regs. went into effect, and the first unit, I believe, was approved in 2000, early 2000.

Since that time facilities that have been using approved digital units have been following the manufacturer's QC manuals. Each manufacturer, because of their different technologies, has a slightly different or sometimes not only slightly, but more than slightly different quality control set of procedures.

And I think what IOM was suggesting is that a uniform set of quality control procedures be developed, and that they be implemented through regulation once they are, and I know that various groups are working on developing a unified quality control procedure. The American College of Radiology, amongst them, has been working to develop this. I believe that there may be some information from the Akron trial, the DMIST trial that could be a benefit in the future, and I think that the goal of trying to standardize these processes is one that FDA is certainly looking forward to. It would make everybody's life easier if it was one set of procedures that the facility, the inspector, the medical physicist would follow.

So I guess part of the issue that this committee can discuss is some of the difficulties and the different technologies that are involved, and if anybody has any idea now what we might do to encourage a development of a uniform quality control set of procedures.

And I look toward the physicists specifically because they're good at math.

(Laughter.)

DR. WILLIAMS: Yeah, this is Mark Williams.

Well, first of all, I agree 100 percent that things may be a little bit different now than they were in the days when full field systems first arrived. We have more data. The DMIST trial was certainly a good source of data because most of the major FFDM manufacturers were involved in that trial. There were systems from one of them.

So we have data for a large number of different types.

Having said that, I think the point that Dr. Finder raised right at the end is also relevant, which is that we'll have to be careful when we set up these unified guidelines to take into account the fact that of the five different FFDM manufacturers involved, there were five very different technologies involved.

Now, there have been several papers that have been published recently that actually did sort of a systematic analysis that compared the quality control guidance that right now, as Dr. Finder said, is really the MQSA regulations, follow what the manufacturer says. And those papers unanimously demonstrated that right now there is a huge disparity in not only the details of the tests that are recommended, but also in the actual types of tests that are recommended, with in some cases very few detector specific tests, and in many cases the tests that are being recommended, and this is a very logical and understandable thing, are modeled very closely after the existing guidelines for screen film.

So I think that with that in mind, there's certainly a clear call for some sort of a unified approach to quality control for FFDM. I think we have now, well, we're in the process of getting some solid data to establish what tests are relevant and what aren't, and I think the DMIST trial actually identified several tests that probably are not as relevant as they might be and, therefore, could be dropped to simplify the FFDM quality control procedures.

And so I guess in my opinion there's clearly a very strong motivation to do this, and I think that there's no doubt that this is the time to push forward on it.

CHAIRPERSON HENDRICKS: A comment from the audience? Please identify yourself.

MS. BUTLER: Hi. I'm Priscilla Butler with the American College of Radiology.

One of the things Mark had been referring to is an ongoing project at ACR to develop a quality control manual for full field digital. This is under the chairmanship of Martin Yaffe in Toronto, who's not subject to the MQSA regulations right now, but there are a lot of people on the committee who are.

One of the things that I think from the DMIST trial we've learned a lot of information about quality control. The DMIST trial was a research study. All of the facilities were very tightly controlled in terms of the QC that was done there and the attention that was paid to the performance of the equipment.

Martin and his group has taken it one step further to try to come up with a system that is going to be applicable not only to research sites, but also to university, to small community centers that maybe begin doing teleradiology.

So currently we have the technologist section of the manual in a semi-draft form. It's going through as we speak pilot testing, right, Mark?

DR. WILLIAMS: Right.

MS. BUTLER: Right. Okay. And we hope to make some changes to it from the feedback that we get. So that's where we are right now.

Oh, yeah, and once we come out with something, we're going to have to come to this group for an alternative standard to see if it can be implemented under the regs.

PARTICIPANT: (Speaking from an unmiked location.)

MS. BUTLER: It is in draft form, and we are going to be pilot testing it.

DR. BARR: Thank you.

So do we wait for that kind of thing to come out or, Dr. Williams, as you indicate, is there enough information right now to write specific regulations related to digital?

Also, I would like to add in this I think that one thing we've learned from our experience with MQSA is that equipment is, anymore in this day and age, is really not where the problems in mammography lie, and we wrote a whole bunch of equipment regulations.

And is that how we also want to go with digital or do we want to learn from our experience that equipment is probably not where most of the issues lie?

DR. WILLIAMS: Well, I think that to answer your first comment, I think that we're probably not in a position at this moment in time to say these should be the regulations for FFDM. I think we're getting there, and I think that once we do some actual in the clinic evaluation of these draft protocols that Penny mentioned, we'll have a lot better idea.

Because they were called to a large degree from a kind of a super set of the procedures suggested by the manufacturers. That's sort of how DMIST was put together. Everything that was really possible was really done.

And so part of the process that's going on right now is identifying the minimum useful set, if you will. We don't want this to be a big and burdensome set of QC procedures simply because it's going to be applied across the boards.

So I think that probably we should let the ACR subcommittee do a little bit more work. That would be, I think, well worth the wait.

DR. BARR: Thank you.

MS. MOUNT: Carol Mount. I'm just hoping that when these regulations are written, it is taken into consideration the down time for the room to perform the procedure. Currently we have one digital unit and 11 film screen units in our institution, and the digital unit requires so much more down time to do the QC on than the other rooms, and that's time they're not doing a patient.

DR. BARR: Thank you.

DR. MARTIN: Melissa Martin.

I would just reiterate what Mark Williams has been saying and Penny. The ACR group has put a lot of work into this already as far as trying to develop a cohesive set of requirements that we would recommend at that point to be included particularly for the physicist test and the technologist test.

I hesitate to have the FDA start or recommend that you start developing a different set at this point. I really would encourage us to wait until the ACR program gets out because that's the group that's been working on this, and the accreditation program is fairly advanced at this point rather than starting another set of criteria.

DR. BARR: Thank you.

DR. WILLIAMS: Just to comment on the issue of the down time for the digital rooms, that's probably true at the moment, and part of that has to do with the, I guess, unwieldiness of what it is we're all trying to do when we're in there doing these tests.

One of the things that, of course, we all hope is going to be a benefit of digital, in addition to its clinical value, is being able to computerize many of the things that right now are done in maybe not the most efficient way, and that might be another virtue of a standardized set of tests, is that if these things could be essentially incorporated up front into the FFDM systems, then it may actually decrease the down time because there would be a well established set of analysis, routines, for example, for doing the test, and we wouldn't be shuffling the images back and forth from one place to another and getting them off and doing off-line evaluations and so on.

So hopefully that will be one of the benefits that will accrue.

DR. BARR: All right. Thank you.

Those are very helpful comments. So what I'm hearing, I think, today is that FDA should continue to require the use of manufacturer's QC manual and check that folks have their initial training in modality, and that there is hopefully imminent information that will, although not today, soon allow us to write a specific set of regulations of the necessary elements for digital units.

DR. MARTIN: Can you clarify? I guess I'm asking for a time frame clarification. What will it entail? In other words, if this program were released from ACR and implemented arbitrarily January 2006, which it's not going to be ready, but say it's ready January 2006. What is the time frame for FDA then to adopt that so that the facilities are not caught in the requirement to have an ACR program and an FDA program?

Can you elaborate a little bit on how that's going to work?

DR. BARR: Yeah, I think Dr. Finder, who is our regulations expert can probably help with those time frames.

DR. FINDER: Now, is that January 1st?

(Laughter.)

DR. MARTIN: Fifteenth.

DR. FINDER: Fifteenth. Okay. Well, assuming it comes out on the 15th and it's not a weekend, there are two different aspects to it. One is the issue that Ms. Butler brought up about an alternative standard. They could submit something to us. We would review it as an alternative standard. Those usually go through within a matter of weeks or months in order to get those through the process.

And what that would allow, it would allow facilities to use that standard instead of what the manufacturer recommended. If you're talking about regulations such that this now would become the only de facto standard, then you're talking about going through I notice a common process that would probably go anywhere from 12 months to 18 months, probably more on the 18 months side, and would have to come before this committee and go through a formal process.

The alternative standard process can be done within the division because, as I had mentioned earlier in the morning session, we do have the ability, the authority to grant alternatives if those qualifications that I mentioned are met. And this might be one of those in which there were sufficient data to show that this would improve quality, speed things up, and could be approved through that process.

DR. BARR: Thank you.

For the physicists on the panel or Ms. Butler, anyone who wants to comment, do you feel that if we get to these tests that we feel are necessary, that the technology -- how do I say this -- will stay stable enough for a while that these will be, you know, implementable, or are we in such a flux right now that we're going to be looking at approving alternative standards or changing regulations constantly to accommodate digital?

DR. MARTIN: I think for all of us concerned we hope it's stable enough that whatever is developed would be adaptable to any of the new technologies that are coming on.

DR. BARR: Thank you.

DR. WILLIAMS: Yeah, I'd say there's probably no good way to predict what new technologies might arrive on the scene, but presumably this would be sort of a self-equilibrating thing. If there were a set of well established standards for performance, then in the FDA approval process for the instrument, then hopefully some of these things would get ironed out.

DR. BARR: And you think we could make this adaptable to the technology?

Penny, did you have?

MS. BUTLER: The group has been trying to write the tests general enough to accommodate the different technologies that are out there now, and there are going to have to be different specific procedures, which is going to be specific for each manufacturer just because of the way the equipment works.

Be that as it may, currently working with the manufacturers and their own QC manuals, we've gone through numbers of different revisions of their QC manuals for the same model of equipment based on software and everything else.

So I think it would be worthwhile once we hit the regulatory stage to try to build in some creativity to allow for some changes as we go along. Hopefully we'll learn more from the pilot testing that we're doing now and we can provide some advice as a result of that.

Maybe some of the manufacturers might have some input.

DR. BARR: Great. Thank you very much.

CHAIRPERSON HENDRICKS: Another comment from the audience. Please identify yourself.

DR. BARR: I was looking around for him.

DR. SANDRIK: John Sandrik, GE Health Care. We manufacture and sell medical equipment.

I think the idea that this will be stable is wishful thinking. I know certainly in our own equipment at least two developments that will affect QCs under PMA submissions right now. There's been presentations to this group. A couple of years ago Dr. Kopans was into the total synthesis systems. Certainly that's going to induce a lot of entirely new QC concerns, but essentially it's an outgrowth of a digital mammography system.

I guess he agrees.

(Laughter.)

DR. SANDRIK: But I know as Penny has mentioned -- we must have lost him. Should I continue or do we need to have a quorum in place?

Well, anyway, as Penny brought up, we've gone through several issues with our QC manual. We have software changes,. We have hardware changes, and we anticipate that those are going to continue.

I think one big difference between sort of the evolution of screen film and digital is that the mammography community had ten to 15 years of experience since green film before we even thought about setting down regulations for how it should be quality tested and evaluated.

You've had at most five years on one system and probably only one or two years on some of the other systems, and there's things you haven't even seen yet. So there's just not that kind of experience.

And I guess one thing, if I would add one plea here, you know, I think Dr. Barr had mentioned something before earlier. Having data before regulations. You know, I think I've looked at at least an outline of what the ACR has presented. We looked at what many of the manufacturers have, and I think there's a lot of consensus on what tests to do and there's probably even some consensus on what tests we're wasting our time on, but they're there because they're part of regulations or whatever.

But I think the big problem is setting what the action limits or the upper or lower bounds or whatever they are, the limits of acceptability. And that's a place where I have concern in terms of having the right data in order to make those limits relevant, and I know I have had some discussion with some of the DMIST participants on whether that data from the DMIST study could be used in QC development.

And at the time the response was that the study was never set up to do that. You know, so the ability to take some of that data and work it into QC limits may be something that still has to be worked out, but at least that could be some sort of source of information.

But I would really like to see some of the things like we were talking about with interpretive skills and all of the other applied to QC, that there be some sort of data to say, yes, sending it at this level really is going to make a difference between mammography quality, and it's not just a number pulled out of the air.

Thank you.

DR. BARR: Thank you.

For our transcriptionist purposes, I don't think this was ever said that DMIST is digital mammography imaging screening trial.

CHAIRPERSON HENDRICKS: Another comment from an audience member?

MR. UZENOFF: My name is Bob Uzenoff, and I'm with Fuji Film Medical Systems, and we have a digital mammography system that's currently being reviewed in the FDA as a PMA.

And I would like to point out I think the wisdom in the original MQSA act of allowing for innovation and new technologies, the technology in our system which is under review is not the same exactly as devices that have been approved already. It was part of the DMIST trial, and so there is experience with that quality control program clinical experience, and I think the kinds of tests, as the previous speaker mentioned, we have an idea, I think a pretty good idea in physics of the types of things to look at, but just literally looking at the recommendation, they are a uniform set of quality tests and test criteria is a little strict.

Dr. Finder's recommendation of the alternative quality standards, I think, would nicely accommodate evolution in technologies and accommodate various technologies. In this X-ray realm, things are done differently, but it's not totally new. It's not like the difference between X-ray and MR.

We know about subject contrasts. We know things about resolution. We know about noise. We know what's important, but how to measure them and to set criteria, I think you'll find it's a little early to do that.

Thank you.

DR. BARR: Thank you.

DR. MARTIN: I was just going to reiterate the fact that, I mean, that's what Dr. Williams and I were saying with the ACR program. The ACR program will be pilot tested because, again, I agree complete with the speakers. We do not want to bring a program; we should not be implementing a program that has not been pilot tested to make sure it will work with all of the manufacturers, and that is the purpose. That's why that program is not out yet.

It's going to be tested before it's brought up.

DR. BARR: Thank you.

Well, I still think my idea of just flipping through this digital section was probably the best thing, but you all stopped me. So you know.

(Laughter.)

DR. BARR: Okay. Thank you for your comments.

And now we get to our green and peach shading and specific regulations. This is Section 912. Overall final assessment of findings are negative. There were no recommendations to change that. There was a recommendation to add the word "finding" or "findings" after "benign," also a negative assessment.

Let's run through these and then we'll go back and see if anybody has any comments on each one.

I see probably benign recommendation to add "finding." Initial short-term follow-up suggested a finding or findings has a high probability of being benign.

Under recommendation for D was suspicious, to add "abnormality biopsy should be considered."

E, to add a biopsy should be considered after "highly suggestive of malignancy."

And new F, to add the wording "known biopsy proven malignancy, appropriate action should be taken. Reserve for lesions identified on the imaging study with biopsy proof of malignancy prior to definitive therapy."

Okay. B, adding the word "finding." I mean, unless -- is there major comment point anyone wants to make?

C, adding "finding," initial short-term follow-up suggested to the probably benign category. Any comment there?

D, under a suspicious, to add "abnormality biopsy should be considered." Anything there?

E, "highly suggestive of malignancy, biopsy should be considered." Any comment?

And then F, the known biopsy proven malignancy, and this says, "Reserved for lesions identified on the imaging study with biopsy proof of malignancy prior to definitive therapy." I guess I myself would wonder, you know, what about during definitive therapy. What about, you know, immediately following definitive therapy? Any comments on F?

Okay. In cases where no final assessment category can be assigned due to incomplete work-up, incomplete needs additional imaging valuation. The recommendation is to add "and/or prior mammograms for comparison."

"Show the assignment as assessment and reasons why no assessment can be made shall be stated by the interpreting physician, and a recommendation to add for cases rated zero because of need for prior examinations, reassessment must be performed within 30 days to assigned category."

Any comments here?

MS. PURA: Dr. Barr, Linda Pura.

How come we don't just go right now the BI-RADS and use the BI-RADS as opposed to the various categories that are medically reported? I think the docs get very confused with they're reported in the BI-RADS. Why can we not just use the BI-RAD category one to zero to six instead of the alphabet?

I mean, it's not a major point, but I know a lot of our docs get very confused with those. It sounds very basic, but it's very true in practice.

DR. BARR: Well, maybe we should just get rid of BI-RADS and start over. How about that?

Charlie, do you want to comment?

DR. FINDER: Well, yeah. Let me go back to a little bit of history and kind of put some of this into perspective and where some of this is coming from.

The goal originally was to create a system so that the referring physicians would understand what the reports basically said. Before this requirement went into effect, reports could be long descriptions of things without any assessment whatsoever.

When we put into regulation the assessment categories, we basically picked the wording from the BI-RAD system. We basically used that. At that time, there was some discussion about using the numbers, and the feeling of the committee at that time was that the numbers themselves were not sufficient because then there would be confusion about what the numbers meant.

So what we did was we said you have to sue the language of the assessment categories. If you wanted, you could add a number with it, but the wording had to be there. And as I say, we basically took the wording from BI-RADS.

Now, over the course of years, things have -- we've learned. Let's put it that way, and not only we have learned, but BI-RADS have learned, and some of these have been modified to take that into account, not always with the best of results, and I'll give you one example in a minute.

But in addition to the language that was in the regulation for these assessment categories, we found that some facilities were using slightly different words, and what was happening was we finally had enough problems with that, enough facilities were being cited that we came up with a list of equivalence, and that's in our guidance, other wording that we would accept as equivalent to the assessment categories.

What is now happening is those lists are enlarging, and it's now getting to the point where you can pretty much write almost anything -- well, I shouldn't say that. It's not that bad, but it's getting confusing enough so that facilities are now having problems even understanding the BI-RAD system because the latest BI-RAD system now has broken down some of these categories even further, and you've now got things like low suspicion, moderate suspicion. It's getting more and more confusing.

One of the problems that we've done, and we've actually accepted the alternative standard for this one is the one that talks about the incomplete category where we've added and/or prior mammograms for comparison.

We've gotten feedback from some facilities and from some referring physicians that now they don't know what this means anymore because in the old days it would just be incomplete and need additional imaging evaluation.

With the current wording, now they don't know whether the patient needs additional imaging evaluation or they're waiting for a comparison. So in order to be helpful, in order to be flexible, we may have created a system that is even more confusing.

Another difference between our assessment categories and BI-RADS, as at least written here is we did not tie the assessment category to a recommendation. We gave the facilities flexibility to use an assessment category and supply a different recommendation if they believed that was indicated.

If we make this a regulatory change, then that won't be allowed. Okay? Some of these will now be tied to specific recommendations. So those are things to consider with this.

Another difference is that while IOM recommended that we add one of the approved alternative standard assessment categories, number F here or letter F here, they did not deal with one of the other ones that we had already approved, and that deals with marker placement during an interventional procedure. Why they didn't include that I'm not exactly sure.

And those were the comments that I wanted to make before you guys started discussing these suggested changes. So that's where we basically came from.

The whole idea of this is to make it as clear as possible to the referring physician what the interpreting physician thought of this mammogram and what should be done next.

DR. BARR: And I really wasn't being flip when I said maybe we should ditch BI-RADS and start over. I was trying to get to what Dr. Finder was saying, which was the original intent. I think we've come full circle now, and maybe the radiologist speaking his or her intent into the dictation of what should be done with this patient and the results of the mammogram is a viable alternative to keeping adding onto categories and allowing more variations of words, et cetera, et cetera.

So I'd like to hear your comments.

DR. FINDER: One other issue that has come up is that the assessment categories here basically are an assessment or some kind of graduation or quantification of malignant status, how malignant you think this mammogram represents.

We have had a case that's been brought to our attention where a ruptured implant got an assessment category of negative because there was no evidence of malignancy, no suspicion of malignancy, and that is cases going to the courts now because they got a negative assessment with a ruptured implant, and they're bringing that as an issue.

So I think there's some confusion as to what the purpose of these assessment categories are supposed to be, whether they only refer to malignancy, whether they refer to even benign conditions of the breast.

We've always had complaints or comments about these assessment categories don't necessarily fit male breast mammograms, and that it's not appropriate for that.

So I just want to hear your comments, your thoughts. Should we be looking at a new assessment category or should we try and define the old one?

I would also state that most of the facilities, the vast majority of the facilities are familiar with this, and to change this would be a huge change in the way mammography facilities practice. So we have to be very, very careful before we suggest anything.

I also see a hand going up, and I can answer this question. Any change that we made in these assessment categories would cause a huge change in software companies that have to redo all of their programs.

DR. BARR: From the audience.

MS. MYERS: Susanne Myers, again, with Mammologics.

One thing I just want to point out when you're looking at these. A lot of facilities -- and I think Dr. Finder was alluding to that -- they tie their patient notification letter messages to these categories, and so if you make any changes to that, it's going to be a challenge for the facilities to get the correct letters to the patients. That could even cause more confusion.

So just something to consider.

DR. BARR: Thank you.

Yes, additional comments from the audience?

MS. BUTLER: Penny Butler from ACR.

If could move to the previous slide.

DR. BARR: Sure. Maybe. I don't know.

(Laughter.)

DR. BARR: Charlie, can you help me?

MS. BUTLER: I just wanted to point out that number E there, which is BI-RADS Category 5, I believe the BI-RADS describes "appropriate action should be taken," not "biopsy should be considered."

DR. BARR: Yeah. But do you think that this is the IOM recommendation, or do you think we have the IOM recommendation wrong?

MS. BUTLER: I don't know. I would have to look at my IOM book.

DR. BARR: Yeah, I didn't know if you were intimately involved in that.

CHAIRPERSON HENDRICKS: Additional comment from the audience?

DR. BASSETT: I think that the BI-RADS has become not only national standard, but an international standard. Most of the countries that have developed accreditation programs have developed and incorporated this into their programs. It works, and I think it's a mistake to change it.

When you give someone an F for what's really a five, basically you don't have to tell the surgeon what to do. They may have different options they want to take. They may want to take the patient directly to surgery in certain circumstances, depending on the clinical factors and so on, and that's why it was called "appropriate action should be taken."

And actually that's used for F, but it should also be for E, highly suggestive of malignancy.

DR. BARR: Yeah, I think that's what Penny has.

DR. BASSETT: Almost 100 percent sure, and again, I'd emphasize what was spoken before, that many facilities have tied their auditing and so on to these numbers, and they've got it in their software and so on. And we're trying to encourage them to do audits, and yet we're going to make it even harder for them because all of their previous studies are identified by the numbering system.

DR. BARR: Would you have a recommendation for the one that said to put in "and/or prior mammograms," "additional imaging and/or prior mammograms"? Because clinicians have told us they don't know what they're supposed to do. Does that mean the radiology department is taking care of it? Does that mean they have to do something?

DR. BASSETT: No, it's a very difficult issue. It has to be clear that what you're asking for is old films versus more imaging. For several reasons it should be identified that way.

One is that you want to keep track of them. So our quality assurance person would have to know which were old films because she would be pursuing those until they are found or let us know if she couldn't get them within a certain time.

And the ones with additional imaging are actually going to count as call-backs in your medical audit. The old films really don't need to be call-backs.

DR. BARR: Would you have a comment on the rated zero, giving a time frame for the zero rating? I think I have it up on the screen now for cases rated zero. A time frame for changing the assessment.

DR. BASSETT: Oh, I think that's a goal to go for certainly in terms of if you can contact the patient and find the patient and so on to get them back.

I think that's not inappropriate, and justify if you can't do it.

DR. BARR: Thank you.

DR. MONTICCIOLO: I agree with Dr. Bassett that that is a goal, the 30 days, but to make it must be performed in 30 days, I mean, we have patients who think that if they get their mammogram right before they go on vacation, somehow they'll be saved from anything bad because, after all, everyone knows they're going on vacation.

And so we have this continual problem with little ladies who are going off for a month or two and have their mammogram right before, and then we try to get them back, and they say, "But can't I come six weeks from now?" or whatever.

And so we have a hard time trying to track them and get them to comply with the regulations when they aren't aware of them. So I think it's a good idea to say, gee, every effort should be made, but to make it mandatory that it be done within 30 days, I think it sometimes is difficult.

I mean, we call them back right away, but there are just some patients that won't comply with it.

DR. FINDER: It's Dr. Finder.

I just want to clarify one thing. The way this is written, the recommendation, it's only those in which you're waiting for comparison films, not in which you're asking for additional studies that would have to be redone in 30 days.

So it's only --

DR. MONTICCIOLO: Well, that's even a bigger problem for us, getting a facility to send an old film. We have to put a tremendous amount of resources into that. First of all, we often send letters, faxes, calls, and if it's outside our general area, it takes weeks and weeks to get these films. We have films show up six weeks later all the time. So the question is what to do with those.

DR. FINDER: Well, that is the point because the impetus for this recommendation from IOM is the situation where somebody reads a mammogram as incomplete, needs comparison films, and send out that report.

And right now under the current regulations, there is no requirement that a, quote, unquote, final assessment category go out at some point in the future. This is an attempt to require that that happen. So if you don't get those comparison films, you will reassess those films and give an assessment based on what you have at that point.

And they are saying 30 days. Is that a reasonable time frame? Isn't it? But it's to address that issue and to prevent people from sending out incomplete studies and never getting the comparison films and never giving a final report, in effect.

So that's the issue that's really being addressed here.

DR. FERGUSON: Don't we already have a 30-day requirement?

DR. FINDER: The requirement is that a report has to go out in 30 days, but if that report is a zero, an incomplete, that has met the requirement. So that it is possible for somebody to send out that report and then never get the old films and never have a final report go out.

So that was their attempt to address in regulation that issue.

DR. BARR: So -- go ahead.

DR. FERGUSON: I was going to say I guess I was under the wrong assumption that you had to do something in 30 days. So what we do at 30 days if we don't have the films, we send out a report, and I guess that you're wanting to mandate that that be done. I thought it already was. I guess I was --

DR. BARR: This is an IOM recommendation, and if we keep BI-RADS because it's standard and because it's attached to patient notification, so in B do we accept a recommendation of adding a word "finding" because somebody thinks it should be added?

Do we keep giving alternative wording? And how do we --

DR. FERGUSON: I think we leave the words alone. I think that we're pretty standard, and that's what we've looked for for a long time is standardization. I think it's everything we've talked about. So I'd say leave it alone.

CHAIRPERSON HENDRICKS: Another question from the audience?

DR. LEE: I have a comment. Carol Lee from ACR.

I agree with what Ms. Pura said. I want to emphasize how our clinicians now having been living with BI-RADS for the length of time that it has been in existence now have a really good understanding of what the numbers mean, and to change at this point, I think, would introduce a lot of unnecessary confusion into an already confusing area.

The other comment I wanted to make is that the BI-RADS committee of the American College of Radiology has devoted an incredible amount of time and energy in developing the wording, and lots of effort by experts has gone into this, and I would urge FDA in their regulations to keep the same terminology as BI-RADS so that we're all talking one language.

Thanks.

DR. MONTICCIOLO: Carol, can I ask you a question?

DR. LEE: Sure.

DR. MONTICCIOLO: This is going to add something that's not in this, but I just want to ask since you're familiar with the BI-RADS committee, what about the patient who has a palpable abnormality but no mammographic findings? Because in my --

DR. LEE: This is something that confuses clinicians terribly.

DR. MONTICCIOLO: Absolutely, and my understanding, and actually Dr. Bassett is a longstanding member of the BI-RADS committee; my understanding is that the committee is addressing that question and others, for example, the implant, you know, the cases that are not suspicious for malignancy, but have other findings.

And the BI-RADS committee is not a one shot thing. It meets on a regular basis, and it is addressing these issues that come up, and that's why I would urge that that language in BI-RADS be adopted.

DR. BASSETT: That's another issue that ‑-

CHAIRPERSON HENDRICKS: Please reintroduce yourself every time you come to the microphone.

DR. BASSETT: Larry Bassett representing Society of Breast Imaging.

CHAIRPERSON HENDRICKS: Thank you.

DR. BASSETT: One other reason that you might want to not jump into that area is because it has now been extended to ultrasound and MR in a way that whenever possible it is kept the same. Like if a mass is round on mammography, ultrasound and MR, it will have that same terminology.

So when you start messing with the mammography when you're also going to affect the ultrasound and the MR, which were developed by people who spent a long time getting consensus because everybody was doing it differently, and the same for ultrasound. They now have the standardized terminology, and that's another big step forward.

So this is not just related to mammography. It's every imaging modality in breast imaging, and it was made to be flexible. So if there is a reasonable reason to change it, the committee will change it. They just need the input.

And it has changed over the years. For example, Category F, which we call Category 6. Category 6 was added because so many patients are getting induction chemotherapy prior to having their definitive treatment with surgery, and those patients need to be evaluated with imaging a lot of times, and so to give them a four or five would be appropriate. So we give them a six, and that's where that came from.

So that's another one that will be in all three types of modalities.

So I think it's a mistake to change something that has taken so long to develop, and it has finally gotten national approval. It is flexible though, and it can change for those ways.

For example, Category 4, which is suspicious, there's an option now to make it 4(a), (b) or (c) because it's such a wide category. So if it's just slightly suspicious, most likely a fibroid (unintelligible) cyst, why not do a cyst aspiration? That's a 4(a).

And then if it's just intermediate in its suspicion, it would be a (b). And if it's higher suspicion like 50 percent and above, then it would be a (c), and then five is restricted for those that you'd bet your house on it basically.

CHAIRPERSON HENDRICKS: Does the committee right now perceive that there's a problem with the Category 0, recognizing the two separate sets of patients.

DR. BASSETT: They've had a lot of comment and work on Category 0, absolutely, and they're working that out, but they're trying to get some consensus and input from all the other societies as well, not just radiology, but surgery and so on. So it's a difficult process.

But you can subcategorize zero into zero-zero if you want to do that here for old films, zero old for old films and zero (a) for additional imaging, too. That's another option that you can use.

CHAIRPERSON HENDRICKS: I see. Thank you.

DR. BARR: Thank you very much.

I just wanted to show the rationale for these recommendations BI-RADS categories to minimize confusion between interpreting physicians and other clinicians, and that FDA has already approved the new Category F in an alternative standard.

Thank you. Those are helpful comments.

D is the establish luminance standards for viewing mammograms and the proposed wording to the appropriate regulatory section is viewboxes used for interpreting mammograms and clinical image quality reviewed by the technologist should be capable of producing the luminance of at least 3,000 candela per square meter. The illumination levels must be less than or equal to 21 lux.

The committee says that evaluation of viewboxes during inspection is not recommended. The rationale is viewing conditions are critical to detect subtle contrast differences, and that the 1999 ACR quality control manual has suggested standards.

The one comment I would make is the standard comment I have on dealing with regulations, is it's important enough to put in a regulation, but it's not important enough to have an enforcement tool for it, and that's always a problem when you recommend putting something in regulation and then there's no way to enforce it. The recommendation, the evaluation is not recommended during inspection. So it's a regulation that we can't enforce if we don't have a compensatory inspection or enforcement component.

DR. MARTIN: Melissa Martin.

I'm confused when you say there's no inspection because the physicists do this. As far as those of us who are inspecting ACR accredited facilities, I guess I would highly recommend that this be approved because we're making this measurement on an annual basis as part of our annual physics report already.

DR. BARR: Right, but it's not part of the inspection procedure, and IOM doesn't think that it should be.

DR. MARTIN: I beg to disagree with IOM.

DR. BARR: Thank you.

Any comments on this standard?

CHAIRPERSON HENDRICKS: From the audience.

MR. MOURAD: Wally Mourad, FDA.

It's true that it's not in the inspection procedures, but if it's in a physicist report and if the physicist says it's wrong, fix it, the facility has to fix it. So in a way it's inspected.

DR. MARTIN: Well, it's inspected and it's part of the physicist report for those that are ACR accredited, but again, it is a measurement we are making. We can recommend, but it would be a lot more forceful if it were part that they had to fix it because right now it's only a recommendation. That is true. They do not have to fix it.

We can tell them all day, but there's no teeth to it.

DR. BARR: Exactly. Thank you.

From the audience?

MS. BUTLER: Penny Butler with the ACR.

One thing that AB's accreditation bodies look for during the three-year accreditation is that we get a copy of the physicist report. If the physicist says that a certain regulation is not met, we will not accredit them until we get something back from the facility saying that they have corrected the problem. So in that sense it is in force when they go through accreditation.

CHAIRPERSON HENDRICKS: What is your take on this recommendation that the viewboxes not be evaluated? I'm just having a little trouble understanding the background for this IOM recommendation that the viewboxes not be inspected.

MS. BUTLER: Not be evaluated during annual MQSA inspection. I agree with that, and I agree with that because it would be checked during the medical physicist annual survey, and so there would be a measurement to determine if it does meet requirements. There would be oversight by the accrediting body to make sure that it meets MQSA requirements.

CHAIRPERSON HENDRICKS: You feel their intent might be that it was a duplication of something that's already in place?

MS. BUTLER: Yes.

CHAIRPERSON HENDRICKS: I see. Is that also your understanding, Dr. Barr?

DR. BARR: I'm interested in knowing how it can go from what Ms. Martin says, which is a recommendation by the physicist that if this doesn't meet, that it be fixed to something that, you know, if it's that important, it needs teeth.

So I'm a little confused about the recommendation.

DR. MONTICCIOLO: I think the issue here is time during inspection because this would take extra time in the inspection, and as Penny Butler pointed out, it already is required to be fixed by the accrediting bodies and so there is some teeth in it, and I know that to be the case because sites that I have checked when they had this problem, the ACR's hand in it forced it to be fixed, based on the physicist report.

DR. BARR: And that's an every three year process, the accreditation. I just wanted to point that out.

Thank you.

DR. MARTIN: I would reiterate I'm not in any way saying that this should be done by the MQSA inspector during their annual inspection. It is something to be handled by the physicist. I would just recommend that, you know, if necessary, this body recommends that FDA adopt that as a standard, but I'm not endorsing at all that it be part of the MQSA inspector's task. This is a physicist task.

DR. BARR: Thank you.

DR. FINDER: Yes. Dr. Finder.

I just wanted to kind of go back and give some history about this issue because when the final regulations were being worked on, the issue about luminance standards for viewboxes was discussed. In fact, viewing conditions in general were discussed. It was decided at that point not to mandate high luminance viewboxes for mammography.

Instead, what the recommendation from the committee was is to use or require hot lights to be available which can produce these levels of luminance without having the more expensive viewboxes.

There was an issue about masking, and that I think is an issue that should also be considered if you're going to talk about the viewboxes because some testimony we got was that if you don't mask appropriately on these higher luminance viewboxes, it can actually worsen your visualization of the image because you're getting all of this extraneous light hitting your eye.

So I wouldn't necessarily just limit it to the viewbox. You might want to also consider viewing conditions. I will tell you at the last time this was discussed we got into the issue about practice of medicine, and people at that committee were hesitant to go too deeply into this. In fact, the recommendation from the committee was not that we require that masking be used; just that the facilities have masking available.

So a lot of these issues probably go into this one thing. I guess the question is do we look at viewing conditions in general and come up with some specifications for the entire range, including use of masking, use of certain types of viewboxes, illumination levels in the room itself which are mentioned in this requirement that they suggest.

So what do people think? How far should we go on this and is this an area that we should be getting into again?

DR. WILLIAMS: Don't we already have recommendations in ACR guidelines for two out of those four things that you mentioned for the background light that's hitting the monitors? So the illuminance and the luminance of the monitors themselves.

As far as masking goes, probably for soft copy viewing it may not be quite as much of an issue since you don't have the bright borders to worry about, and I forget what the fourth one was.

DR. MARTIN: No, they're all in the ACR. Basically the question, if I understand it, Dr. Finder, you're wanting us to -- are you wanting to know if the committee wants to recommend that MQSA or that we recommend the adoption of what's in basically Test 11, the viewing conditions for the ACR manual at this time?

Because all of those items are covered.

DR. BARR: Right, and I think that's the question, is this something -- are viewing conditions, including the luminance and lots of other things related to viewing conditions, something that we want to be into and regulating?

DR. FINDER: Right. Another issue to keep in mind that have been previously brought up before, that over the years the optical density of the films has increased so that there are darker films. So that increased luminescence or illuminant viewboxes might make more sense now than they would have, let's say, five or ten years ago when we were talking about some of the initial regulations.

So, again, just we want to hear your opinion on whether we should go ahead with further regulation of viewing conditions.

DR. BARR: And if we put in regulations that you have to mask, how do we enforce that? You know, does the inspector watch the radiologist read?

I mean, you have to think of when we do these things how do we go about making sure that they're done, or do we?

DR. FINDER: And I would also add to that the issue of does anybody have any idea about how many viewboxes would not meet these conditions and how many facilities would have to get new viewboxes and whether you could achieve the same result using a hot light versus this.

MS. RINELLA: Let me just add, I'm Diane Rinella, a mammography consultant.

I travel throughout the United States. I've been all across this country, and the majority of the places that I do work at, I'm working with them on actual patients and viewing films on their viewboxes that they're using for their criteria image critique.

And the majority of these viewboxes when I ask the technologist are these the same luminance as your radiologists, they look at me with a blank face. They have no clue. They do not have hot lights. They don't have masking, and their overhead lights are on, and they don't know really that these are not the way to do films.

So I'm glad you brought this up.

DR. FINDER: Are these the techs or the interpreting physician viewboxes?

MS. RINELLA: These are the technologist's viewboxes that should have basically the same luminance as the radiologist reading the film.

DR. FINDER: Okay, because the regulations deal with the interpreting physician's viewboxes, but we certainly recommend that they be similar to.

DR. BARR: Well, this recommendation is for review by the technologist also.

DR. FINDER: Right. I'm just bringing up what currently exists.

MS. RINELLA: But like what you said. How are you going to keep the technologist from turning on the overhead lights when they're actually reading films or picking up a magnifying glass and masking their films?

You know, you really can't stand there and be the mammo police, but I think the more aware I think that they are going to be made of this if we do mandate something, I think that could only help.

DR. BARR: And there is an argument to be made that when you put something in regulation, even though you can't enforce it, you know, it obviously carries more weight, but I want people to realize our limitations on some of these things that are recommended for regulation.

CHAIRPERSON HENDRICKS: I would welcome input from the accrediting bodies as to whether these -- you know, where poor viewing conditions is a reason for failure to accredit a facility and how often that occurs. Are there instances where facilities have failed on these viewing conditions that we're discussing? And how often is that a significant issue.

Please, Penny.

MS. BUTLER: Penny Butler from the American College of Radiology.

Currently we don't fail anybody for this because it's not a regulatory requirement. Our standards for accreditation have to be essentially the same as the MQSA requirements. So even though we have it as a recommendation, it's a recommendation.

CHAIRPERSON HENDRICKS: Thank you for your comment.

DR. BARR: Thank you.

DR. MARTIN: Dr. Barr.

DR. BARR: Yes.

DR. MARTIN: Melissa Martin.

I'm sort of like Diane. We consult all over the place. I would say at this point about 20 percent of our facilities would have to replace their light boxes. A good probably 75 to 80 percent of them are already in compliance, and I guess I'm surprised at your question of how would it be inspected because my understanding is every year the local MQSA inspector is asking the facilities to demonstrate how they mask to interpret their phantom films at this point.

Maybe that's just a local we have very aggressive inspectors, but my understanding is they ask every one of our facilities to show how they're viewing the mammography films and how do they mask off, and they want to see the brightness. They're not making measurements, but they are definitely looking at the viewing conditions every time they come into a facility.

DR. BARR: Well, do you want to comment on that?

DR. MARTIN: -- not an FDA?

DR. MOURAD: No, those are aggressive as you say inspections.

(Laughter.)

DR. BARR: This is Dr. Mourad from FDA.

DR. MOURAD: We tell them not to specifically look for those because we don't have inspection questions for them, but we also tell them if you see something totally abnormal and missing at the facility, you should bring it to their attention. Now, some of them are more zealous than others.

DR. BARR: Thank you.

Okay. So we currently don't have -- we have the physicist report. We don't have any current inspection procedures to deal with viewbox luminance. Do we think that this luminance -- do we think the numbers in this recommendation make sense, physicists?

DR. MARTIN: The numbers are fine with me.

DR. BARR: Thank you.

DR. FERGUSON: It's aggressive. Being a radiologist, I don't know numbers. So I don't know what these numbers mean literally.

(Laughter.)

DR. BARR: Well, that's why I had to ask the physicists. I'm in your boat.

DR. FERGUSON: I mean, I'd like to see in the room what kind of luminance we're talking about as far as background light to know. I wouldn't know looking at how many candela per square meter.

DR. BARR: Yes. Ms. Martin.

DR. MARTIN: If you're reading in a normal, good radiologist facility, you are nowhere close to violating these numbers. You probably are sitting in somewhere with less than six for your local -- your room luminance, illuminance.

DR. BARR: What I was glad to see is that at least there's some idea of paying attention to this, and particularly for the technologists, not just the physicians. I thought that was at least an advance.

I think we have another audience.

MS. SPRINKLE-VINCENT: Hello. I'm Susan Sprinkle-Vincent. I'm a mammography technologist and consultant from Houston, Texas.

I travel also like Diane all over the country training technologists, do the 40-hour initial training in Houston, do lots of hands-on positioning, accreditation assistance, and myself, like Diane, find most facilities that I go to the technologists do not have appropriate viewing conditions.

I struggle with that a lot, especially working with them to improve their positioning skills and their technical factors, and find it a lot of times pretty impossible to do in the conditions that they are given to review their films in.

And then a lot of times unable to get into the radiologist area to use their viewing conditions because they're busy and they're tied up.

A lot of the technologists would love to see this in force so that they would be allowed or their facilities basically be forced to buy them the viewboxes that they need.

Thank you.

DR. BARR: Thank you.

CHAIRPERSON HENDRICKS: We'll take one more question from the audience and then move to the next area of regulations -- thank you -- just in the interest of time.

MR. FLATER: I'm Don Flater with the State of Iowa, and we are an accrediting body, and we're also a certifying group.

And we do have very aggressive inspectors and we require that on every one of our facilities. So it has been done at least in the State of Iowa.

DR. BARR: Thank you.

CHAIRPERSON HENDRICKS: Thank you very much.

DR. BARR: I'm not sure how exactly to summarize this, but I think what I'm hearing is that viewing conditions, not just the luminance of the viewbox are important and should possibly be considered for some regulation in MQSA. I think that's how I'll summarize that for now.

E is eliminate the modality specific CME requirement. The recommendation, if we go to the bottom, is for eliminating the wording "this training shall include at least six Category 1 continuing medical education credits in each mammographic modality used by the interpreting physician in his or her practice.

To perhaps shorten discussion time we are, you know, totally on board with this. We have not been enforcing the modality specific CME requirement and are totally fine with removing it from the regulation. I've heard lots of positive comments on this. So what I'd probably like to limit it to is if anyone sees a major objection to removing this requirement.

(No response.)

DR. BARR: Thank you.

I don't think we need to go through the rationale since I think everybody thinks this is a good idea.

This is 900.4, requiring review physicians for accreditation bodies to specialize in mammography. What IOM would like to have is wording that says at least 50 percent of each year's practice in breast imaging and that the physician be currently actively participating in the modality reviewed at an MQSA certified facility.

I think what IOM is trying to get to here, if I understand it correctly is that physicians who are reviewing films for accreditation should have at least if not more experience in the modalities that they're reviewing, then people at the facilities that they're reviewing for, and I would like to hear if possible just a very brief statement from, say, ACR and, Don Flater, since you're here as an AB what you do require of your physicians looking at modalities, in particular, digital modality.

Please reintroduce yourself for the transcript. Introduce yourself for the purpose of the transcript of this meeting, please.

MS. BUTLER: Penny Butler, American College of Radiology.

CHAIRPERSON HENDRICKS: Thank you.

MS. BUTLER: The ACR requires a reviewer's practice to be in best imaging, I think we say. So --

PARTICIPANT: Modality.

MS. BUTLER: Thank you.

In the modality, yes.

DR. BARR: Okay. So if someone were reviewing for digital accreditation, their practice would be.

MS. BUTLER: I take that convoluted language back then. It would be in breast imaging, but they do have to meet -- if they were reviewing for digital, they would have to meet the MQSA requirements for digital.

DR. BARR: Thank you.

MR. FLATER: That's exactly the same for ours. In fact, we even like to use people that work for Penny to be part of our system.

DR. BARR: Thank you, and that was my understanding of what the accreditation bodies did do.

Does anyone see a problem or have any objections if something were to be added that reviewers had to meet this requirement?

DR. FERGUSON: No, I'd just say that it says "specialize in mammography" and then in quotes down there it says 50 percent. I'm one of those that I do general radiology, but over half of my practice is mammography, and when people ask me do you specialize in mammography, I say no, but I do over half of it. So just so that was clear, you know.

DR. BARR: Thank you.

This is Section 900.4, the results of equipment evaluations. With its initial accreditation application, and IOM would like us to add "the results of," a mammography equipment evaluation that was performed by a medical physicist no earlier than six months, et cetera, et cetera, I don't see a major problem with adding "the results of." Does anyone?

Charlie, do you have a comment on this?

DR. FINDER: No. I just wanted to mention that the next couple of slides really deal with a very specific process, recommendations for changes to the regulations dealing with accreditation bodies, and I wouldn't want to spend too much of the committee's time on going through in detail some of this material. If we can kind of go through it quickly, I think that would be the best thing because many of these changes we've already accomplished through changes in our procedures, and I do think we have some other issues that are more important in terms of facility issues.

So if we can just try and go through them quickly, maybe all at once.

DR. BARR: Yeah, I agree, and some of this is minor wording. So I think we'll just flip through them, and if you see something that you think is a big issue or that you really object to, then chime in.

Again, we have wording in 900.4 to add the words "annual survey" and change the six months to 14 months.

Charlie, any issues here that people might ‑-

DR. FINDER: I think that this works out fine for reaccreditation where under our current situation we allow facilities to have up to 14 months for the annual survey to be done between inspections. It certainly makes sense to allow that similar type time frame for the reaccreditation process so that the facility doesn't have to do two surveys within the same year.

DR. BARR: Okay.

DR. FINDER: So we didn't have any real issue with that. The wording on some of this would have to be crafted so that we don't create a problem with a new facility. We won't allow them to do things 14 months before they actually start in practice.

DR. BARR: Right.

DR. FINDER: Six would work there, but it can be worked out.

DR. BARR: I think the intent here is fairly reasonable.

And this is to delete a section of how the facilities submit their information. Any issues here, Charlie, that need to be brought to the committee's attention?

The bottom line here, I think, is the second bullet. Submission to the accreditation body each year is redundant.

Again, some minor wording changes that I think clarify intent. Charlie, any issues here?

DR. FINDER: No.

DR. BARR: And this is consistent with other suggested changes that the IOM has made.

Again, I think a minor wording change just for clarification purposes, which doesn't change the meaning. This is to make sure that facilities know that all units need to be accredited.

And, again, wording to clarify what facilities must do when they have a new unit. I think that's pretty straightforward.

This is a section in the reinstatement policy to delete some wording. Any comment on that, Charlie?

DR. FINDER: No. Just to clarify that the way it's written it kind of gives the impression that if you reinstate you become a new facility. That is not the case, and by getting rid of those words it would make it clearer.

DR. BARR: In this case they're saying the facility retains its original ID numbers. So we don't want to reinstated facility to be considered a new facility.

This one is change to "continuing experience," and this one might require a little bit of discussion. I'm going to let Charlie sort of lead you with how it ended up this way and what we might do here because I think this one probably would engender a little bit.

DR. FINDER: Yeah. The continuing experience and continuing education requirements for all three personnel categories are written in a similar manner, and they talk about measuring back from the date of the inspection, 24 months or 36 months depending on whether it's experience or education that you're measuring.

And the history behind it is that under the interim regulations, the requirement was that you have to have certain requirements met. It didn't give any specifics of how we were going to inspect against it or measure against it.

And what we were finding was while that I think everybody at the time those regulations were written had the idea that everybody should always meet all of these requirements, the problem that we encountered was that some of our more zealous inspectors were trying to inspect and insure that on every single calendar day somebody met this requirement because you can go back for the last two years and check every single day and see if they met it.

In order to avoid that, we by policy informed the inspectors that they were to measure it in a certain manner, either measuring back from the date of the inspection or from the end of the previous calendar quarter or any day in between. The choice would be left up to the facility.

Well, that was in guidance, and when we rewrote the regulations, it was worked into the regulations themselves, and that's what we have here.

Now, we have gotten many times from many different sources the request that instead of measuring it back from the date of the inspection we go from a calendar date, the first of the year, making it more simple for the facilities to keep this requirement in terms of bookkeeping.

It has been considered multiple times. It was considered before we even put these in the regulations. That concept had to be weighed against the idea of do we want to make sure that everybody always meets all of the requirements all the time.

And we felt that when the final regs. were written, this was a reasonable compromise. In effect, the way these regs. were written and the guidance that's associated with it, if a facility keeps up on a quarterly basis and makes sure all of their people meet their requirements as of at least the quarter, they will never have any problems.

This, however, would make it simpler on the facilities if we went to just a calendar date, and that's one of the things that is being considered.

Another part of this requirement that I think is very important that you need to consider is that they also say continuing experience obtained outside of the U.S. is also acceptable, and we wanted to know what people thought about that.

So there are two aspects to this, and it's very similar in one sense for continuing experience and continuing education. Do we want to change how we inspect against these requirements? Do we want to change it to a calendar date or not? And what do people think?

And once you start talking about that, I can give some more background as to what we have found in the past.

DR. BARR: Okay. So we'll start with the previous two calendar years' additional wording instead of all the current wording that's in here. Any comments on that?

DR. FINDER: Anybody think it should change to what they're recommending, leave it the way it is?

DR. FERGUSON: I don't like the way it is. My technology --

DR. FINDER: You're in the majority then, but the question is what --

DR. FERGUSON: My technologist comes to me every time and says, "We have our inspection coming up. Now we have to go calculate," and we'll spend an hour calculating whether my hours are done according to the way they ought to be done.

And it would be simple and I don't know when you say a calendar date if you're saying January 1st to December 31st.

DR. FINDER: Right.

DR. FERGUSON: I think it would be simpler on the people checking as well to be able to look and say, "Well, in this year you had five, five, and five" or however you did it.

MS. MOUNT: Carol Mount.

I agree 100 percent. It's a nightmare trying to sort back from the date they came for inspection to try to figure out if everybody has the right numbers.

Our numbers in our institution happen to be big enough that we have kind of been doing calendar year anyway, and it works just fine for us.

DR. BARR: Charlie, again, could you just tell us quickly what the reason was for doing it this way in the first place, what the simplicity would not allow for?

DR. FINDER: Right. Some of the advantages and the reason we decided to go with the inspection date is that's when the inspector is there. The inspector can actually see what's going on, can look at the numbers, and if necessary can cite the facility.

If we go on a calendar basis, we would have the following type situation. We would hope that the facility would do what they're supposed to do on the first of January. If they didn't, however, the inspector wouldn't be there. There would be no citation at that point. The inspector could come in 11 months later and find that on January 1st the person didn't meet the requirement.

Of course, at that point it would be questionable whether they should cite or not because by the time the person came in, the inspector came into the facility, that person probably would be up to the requirement or could be, in which case, we didn't want to cite somebody for something that happened 11 months before when they're now qualified.

So it really came down to an issue of could we -- what would be the most efficient way in order to deal with the inspection when the inspector was there, could address the records, and could make the finding at that point rather than leave it up to the facility at some point earlier in the year?

DR. BARR: What if we just used the 24 months from the date of the annual inspection and not the second part of that getting to choose or choosing a quarter?

DR. FINDER: Well, we put that in there to give the facility more flexibility because in many cases while the facility knows approximately when the inspector was going to be there, by allowing them to go back to the end of the previous calendar quarter, they wouldn't have to rush around as soon as they got the phone call from the inspector and say the inspector is going to be here in five days, to start looking at those records then.

They could get their records set as of the previous calendar quarter when they expect the inspection to actually occur. So it was an attempt to make the bookkeeping easier on the facility.

Basically if you're dealing with a single facility what this says is you just have to figure out your numbers once a year, the end of the calendar quarter, before you expect, you know, the inspector is supposed to come in.

The real problem comes up with people that work at multiple facilities where they will be inspected at different times, and that can make it more difficult for them. Each individual facility could do this on a quarterly basis, but the individual would have to keep their records up.

That was not felt by many on the committees in the past to be such a bad thing because their understanding or their idea was that everybody should be qualified every single day anyhow. So it shouldn't be that big a deal for somebody to document that.

But obviously there are other issues, and it's not as simple as if you just pick one date. So you have to weigh those two things.

DR. BARR: Yeah, I think it's pretty clear that the initial wording wasn't just designed to be confusing, that there were reasons.

Any ideas of how to solve this problem?

DR. MARTIN: Melissa Martin.

I don't think it's a solution, but I would reiterate what Carol said. The facilities go through great contortions to meet "oh, the inspector is coming today." The inspectors come within a 14 month period, and at this point maybe because we are a state where apparently there's been a disagreement or not a signoff, and MQSA inspections have been postponed literally at four o'clock on the day before the inspector is supposed to be there the next morning.

The facilities have gone to great lengths to get all of their data together, and they're now told, "We're coming in six weeks," which puts them into a different quarter and they were told to redo all of their data.

Talk about an absolute waste of time. This is what's going on whether we want to acknowledge it or not. It is an absolute waste of time to gather all of this data twice. It would be much more simple just to put it into the calendar years.

DR. MONTICCIOLO: I agree. Those are excellent comments because we've been through exactly that, having to recalculate and recalculate, and we provide our physicians with their audit data on a year to year basis, and so this would match the audit data.

And I would also point out that this type of change, just allowing the use of the calendar year doesn't introduce much danger to the patients or the quality. You know, I don't expect my physicians if they do their CMEs six months later are going to forget everything they knew up to that point.

So I think that if they want to a course three months later or whatever and just stayed in the calendar, they're going to remember it better instead of having us have to badger them because our date is coming up, a date that doesn't relate to anything in their minds and we say, "Hey, you know, you have to get your CMEs."

If they had it yearly stuck on the calendar, it would be easier for them to remember and to accommodate us. So it would be actually a benefit to go to the calendar system.

DR. BARR: Okay, and I think I'll summarize it that way.

I have to agree when I practiced under MQSA I found this confusing, and I still do, but there were reasons in mind when this was set. So summarize here that the current wording is confusing and that we should work on perhaps a calendar year process to simplify.

CHAIRPERSON HENDRICKS: There is a comment from the audience.

MR. MOURAD: Wally Mourad, FDA.

Sorry I did not bring this up earlier. I just thought of it. When we try to calculate the continuing experience and continuing education, it's always referred to the individual starting date. The starting date is the date when the individual has met his or her initial qualifications.

So for people that have met the initial qualifications several years ago, it's not an issue. But for people coming into the fold today, if a person qualified today, they become eligible for meeting the continuing requirements 24 months from today and 36 months from today, respectively.

If you do it on a calendar basis, that equation has to be changed somehow. Just be aware of that.

DR. BARR: Thank you.

Okay. So I think we get the spirit of this, and with our collective brains I think we can work on this issue.

I think the last line of this continuing experience obtained outside of the U.S. is also acceptable. I think we should go through some of the comments on this. I don't know where the rationale is on that.

Well, be that as it may, what do people think of --

DR. FINDER: There it is.

DR. BARR: Oh, physicians who initially qualified in the U.S. under MQSA should not be prevented from using foreign experience. I think some of the things that we've heard as a concern is how do we know what's going on in these facilities.

Charlie, what have you heard on this issue?

DR. FINDER: Well, yes, we have heard concerns about using foreign experience. Under the current guidelines and regulations, foreign experience is not allowed. Our feeling was that we have no idea of what kind of quality is being put forth in those other countries. We have no idea what type of facilities they're at, what type of equipment they're using, whether there's any quality assurance at all being done, whether there's any audit procedures being done.

And we also believe that the two-year interval that we allow continuing experience to be recorded took into account the fact that people may be going out on sabbatical, may be going out for medical reasons, may not be doing a lot for some period of time, as much as a year, and they could still meet our requirements.

We felt that there was enough leeway put in here that the issue about allowing foreign experience wasn't necessary, but that's why it's being brought up before the committee, to hear what you people think.

DR. MONTICCIOLO: Okay. Well, Dr. Monticciolo.

You know, there certainly are a lot of good mammographers in other countries, but I do think it is hard to decide how you're going to gauge quality, and I've done mammography projects in different countries, Panama, China, India, South Korea, and those are all completely different and so I would have a hard time just making a blanket statement to allow it to be used.

Our initial requirements to get somebody to be an interpreting physician I don't think are so onerous that it would prevent people from other countries from practicing. So I don't know. It would be very difficult to gauge those levels of experience.

DR. BARR: Yeah, I'm not sure what prompted this recommendation. I don't know if there's a pressing need to have foreign experience included. On balance, I think I've heard more concerns than positives, but I don't know what prompted this.

Thank you.

Are we on closed facilities already? I'm sorry. Charlie, am I at the end of my thing yet?

This is a change to continuing experience for medical physicists, and what IOM is recommending is to take out the wording in the requirement of at least two mammography facilities and a total of, and I think we can sort of shorten this discussion.

What they're saying is it's difficult to medical physicists to provide services to more than one facility in a 24 month period. Outside consulting is sometimes prohibited. Units' facility ratios increased, and physicists have adequate experience surveying one facility.

I'm not sure what prompted this because our current regulation is that this requirement can be satisfied by the regs. the way our regs. are currently written. So I'm not sure exactly where the issue is here.

If anybody sees where the issue is, I'd be glad to entertain it. Otherwise, any physicist want to comment on this?

DR. MARTIN: I don't think there is an issue.

DR. BARR: I thought it was pretty clear, but obviously it wasn't clear enough.

CHAIRPERSON HENDRICKS: Comment from the audience on that?

MS. BUTLER: Penny Butler from the American College of Radiology.

One thing that prompted this is that there were physicists at large institutions with a large number of units, and because the way the regs. are written, that they provide services at two facilities. Sometimes by their contract they're not allowed to practice outside their own facility. So they would have to really be forced to providing services someplace else in order to meet the regulations.

DR. BARR: But you can't do it by being at a single facility. You can meet the requirements.

DR. FINDER: Right. This is Dr. Finder.

The regulations allow the medical physicist to do a survey. It's a requirement, two facilities and six units over two years. So by doing the same facility twice in the two-year period, which you are allowed to do, one each year for the surveys that are necessary anyhow, you would meet that requirement.

The same for the number of units. We do allow -- for example, a medical physicist who is in a facility that has only one unit can by doing two surveys and resurveying the unit as much as every 60 days can actually meet this requirement at one facility.

MS. BUTLER: Okay. If that is, indeed, the intent, although I don't recall that from the original regulations, and I'm pleased to hear that interpretation; then if that's indeed how it's being interpreted, I think it's --

DR. FINDER: Well, yeah.

MS. BUTLER: -- it's right the way it's written.

DR. FINDER: All right. It's written that way.

CHAIRPERSON HENDRICKS: Dr. Williams.

DR. WILLIAMS: This is Mark Williams.

If that's truly the intent, then why is the word "facility" even brought into it. Wouldn't it be clearer if we just took it out?

DR. FINDER: No. Actually it probably wouldn't because the idea here is that there's a difference between the survey of a unit and the survey of a facility. There are different aspects to both. So if you just did unit surveys, you wouldn't look at the QC for the entire facility, and that's part of what is required for the annual survey.

So that's why it was specifically written referencing two facilities and six units, but they all can be done. It can all be accomplished in as small a facility as a single unit facility.

DR. WILLIAMS: So would there be some merit in explicitly putting that in?

DR. FINDER: It is. It says here, "No more than one survey of a specific facility within a ten month period or a specific unit within a period of 60 days can be counted toward this requirement." That's in the regulation.

We thought it was clear. We also put it in our guidance, too. So it is possible for a physicist who cannot work outside his one facility to still meet the requirements. They don't have to go anyplace else.

DR. BARR: Clearly, we all believe in the spirit of this. We can look at it and see if there's any way the wording can be any different, but the interpretation is as we've said.

Thank you.

The next is changes to the lead interpreting physician requirement, and IOM recommends adding language that the lead interpreting physician must provide regular feedback to technologists on the quality of images. It seems to me that was already their job, but I guess this makes it more explicit.

DR. FINDER: Actually there is a requirement that all interpreting physicians have to give feedback.

DR. FERGUSON: Does that have to be documented in any way? That's the problem you get into.

I mean, we interact every day and say this looks bad, this doesn't, but if you come in and want a piece of paper saying, "Where did you document that you did that?"

(Laughter.)

DR. FINDER: Dr. Ferguson, I know you must tell them this looks good and this looks better.

(Laughter.)

DR. FERGUSON: No, sometimes it's pretty rough.

DR. BARR: Yeah, I don't think there was any requirement that this has to be documented. I don't know if that's what IOM was trying to get at since already this is a requirement. Anyway, I think obviously the spirit is that people should be communicating with facilities to get better mammography.

The rationale here was the on-site surveys. ACR does suggest facilities could benefit from improved physician technologist communication. As my kids would say, "Duh."

And requiring regular feedback may improve quality.

Next is changes to weekly phantom image, quality control test. This would be a change in the optical density of the film at the center of an image of a standard FDA accepted phantom, and it would delete 1.2 and change it to 1.4 when exposed under typical clinical condition.

Yes.

DR. MARTIN: Melissa Martin.

My only comment would be why are you leaving it at 1.4. Shouldn't it be at least 1.5?

DR. BARR: Well, you'll have to ask IOM that. I think the point is well taken about --

DR. MARTIN: One, point, four is way too low.

DR. BARR: I'm not sure why they picked this number.

Anybody else have comment?

So we think it should be at least 1.4 is what I'm hearing.

Okay. Rationale seems pretty logical.

Screen film contact. They want us to take out the word "semi" and put in "annually," and the test shall also be carried out initially for all new cassettes as they are placed in service, and whenever reduced image sharpness is suspected.

The rationale is that this only needs to be performed annually or on new cassettes, and we already have guidance on this that says screen film content tests must be performed on new cassettes prior to clinical use. So I think we're okay here with what IOM intended.

Change to kVp accuracy and reproducibility. They would like us to add facilities with older three phase screen film systems. Take out the end reproducibility part. Take out the most commonly used clinically part, and say that is obtained when the accrediting body phantom is imaged with the mammography X-ray unit set to the most commonly used clinical AEC mode.

And take out the wording in the orange on the kVp and add newer units with medium and high frequency generators will not require this test.

They feel the phrase "most commonly clinically used kVp" is confusing. Data from DMIST shows test really fails during the annual survey. Equipment voltage regulation is tight. Unnecessary on an annual basis.

Anybody have any comments on this area?

DR. MARTIN: Yes. Melissa Martin.

I would agree that I have no problems with that recommendation. The biggest problem we would have is if you tried to enforce the kVp. That does fluctuate on some models, but obviously if you're going to eliminate it for high frequency generators, you just eliminated it on 95 percent of the units. So it's a moot point.

DR. BARR: Okay. System artifact test. They want us to take out tees and target filter combinations and add targets and filters used clinically. The rationale to assess image quality and artifacts, only one test of focal spot size, filter, and target is necessary.

We already have an approved alternative standard for this. So again, I'm not sure what the recommendation is made for, but I think we're okay with this, that we don't have to do all of the combinations.

This is changes to mammography medical outcomes audit. They want us to add facilities with the same interpreting physician should combine medical audit data. I think we kind of covered this earlier in the audit section, that we should allow that combining of data. to make things more meaningful.

And again, they recommended that people not be cited for not doing aggregate data.

Changes to mammography medical outcome audit. The general requirement section they want us to take out "individually and collectively for all interpreting physicians at the facility." I guess that's so we can combine data, and add this whole section here.

A screening exam or a positive exam is defined as incomplete or suspicious or highly suggestive, and diagnostic exams or a positive exam is defined as suspicious abnormality. Biopsy should be considered. And diagnostic exams where positive examination is defined.

Again, this is all rationale for combining data. Not being able to compare facility practice performance with literature.

The BI-RADS committee said the audit of screening examinations requires recommendation for recall, including Category O be considered positive, and it would make the regulations more consistent.

Any comments here, Charlie? Anything that can help a discussion here?

DR. FINDER: Well, we pretty much discussed this earlier, at least some of the aspects of it, and I just want to hear what people think.

I mean, here they have defined screening exams and diagnostic exams, and we have the whole debate about can we do that, should we do that. If we can't define it, it becomes very difficult to write a regulation that talks about those as examinations.

The other thing I found interesting in the way they have it worded -- and I'm not sure if they meant this -- but the way it sounds here is that if you read a diagnostic examination as incomplete, it doesn't have to go into the audit.

Now, maybe I misread, and I doubt it's what they meant.

DR. BARR: Right.

DR. FINDER: But there's nothing that prevents anybody even right now from labeling a diagnostic examination a zero.

DR. BARR: Yeah, I had the same comment on my notes. What about a zero? Isn't that a positive? I thought they already defined that as a positive.

DR. FINDER: And, again, this issue about the screening exams, this would increase the work load for facilities because they'd have to track the incompletes, which could be a fair number, and that also brings back the issue of the incompletes due to comparison with old films versus the additional studies.

So anybody have anymore comments other than what they've already discussed earlier? Is this something that we should put into regulation at this point or should we think about it a little bit more?

CHAIRPERSON HENDRICKS: Comment from the audience? Yes, Dr. Monticciolo first.

DR. MONTICCIOLO: I have to put my glasses on. This is Debbie Monticciolo.

Well, I am reiterating what was said earlier. I do think it would be onerous to have to follow ever zero. I mean, we do ourselves make sure every patient we ask for additional imaging, we make an attempt to get them back and make sure they know they need it.

But to follow all of those to whatever you consider the outcome would be very onerous for any screening site to do, and specially if you include the zeros. So that was actually very well put, Dr. Finder, that I think this is a huge burden, another burden, that we're going to lay on screening sites, and I don't think it would be that productive.

DR. BARR: Okay. Thank you.

Next is Section 900.13, change to FDA action following revocation of accreditation. This is really just a wording change that needs to be done and we agree with.

Okay. Modifying inspections and strengthening enforcement. Under this section IOM said that FDA should eliminate several on-site inspection tests, such as dose and other radiation tests; should require the facilities to cease performing mammography after two consecutive unsuccessful attempts at reaccreditation even if the MQSA certificate is still valid; that we should require a facility that closes or has its certification revoked to notify patients and referring physicians; and regulations for film retention should apply to closed facilities.

So we'll take the first one. Several on-Site inspection tests are redundant and have few failures, and we talked about dose. We haven't had a dose violation since 1997. Dose testing is monitored by ACR and the medical physicist. Other tests are already done by the medical physicist include beam, quality X-ray, film alignment, et cetera.

Again, this just shows you what Mr. Divine showed you about the violations or lack thereof, and we did hear the comment before about possibly looking at the scattering of the dose data around what we currently consider upper limit of normal.

The one thing that didn't come up earlier, I think, related to dose is that you think, well, big deal, the inspector going and modifying it, and if you don't measure dose, is it really going to cut down the inspection time?

But we also have to buy, calibrate, maintain equipment for the inspectors to perform those measurements. So that's another just piece of the pie here.

And the objections I have heard are, as I've said before, about the disparity between the physicist and inspector's measurements, but we're looking at that and I don't expect that to be the case, but we'll see.

FDA should have the authority to require facilities to cease performing mammography after two consecutive unsuccessful attempts at reaccreditation. Our legal eagles here tell us that FDA cannot require facilities to cease mammography if their MQSA certificate has not expired.

Charlie, do you want to lead a little bit of this discussion on this?

DR. FINDER: Right. Again, some of the background on this. Usually this situation occurs when a facility is coming close to the end of its certificate. It's in the reaccreditation process, and it doesn't pass the accreditation process.

Our lawyers have told us that that process deals with the next three year accreditation and certification, not with the current one. What they have told us is that we cannot automatically tell the facility that they must stop doing mammography based on their failure to get a new accreditation.

We can tell them that they should stop. They must stop when their certificate expires, and they have told us that if we feel that they represent a risk to human health, we can take action either to suspend their certificate if necessary.

So what they've told us basically is you cannot take actions against the facility more severe than you would for a suspension, and in most cases when we suspend a facility, we have to give them notice. We have to allow them for a hearing, and they have told us that you cannot just because they don't pass the accreditation process for their next three years automatically tell them to stop where in a situation where the accreditation body tells you that they represent a risk to human health, a much worse situation, you have to give them a legal process to go through.

We certainly have that ability in these situations to take that more extensive action, but the reality is that by the time we could actually suspend a certificate under those types of circumstances, their certificate would have expired anyhow. Usually it's a matter of a few weeks at most, in most cases.

So this is true, and the recommendations from IOM that we put ourselves or grant us the authority in regulation certainly would address this situation. I'm not exactly sure that we could still do it because we cannot put in a regulation that negates the entire appeal process, the entire process for a hearing. So I'm not sure we could even implement it even if we tried to write a regulation.

But that said, I think the concept of is this enough of a problem that we should try to develop some method for dealing with it or is it so self-limited that we should just pretty much leave it the way it is and let the certificate expire, understanding that these are facilities that we have no indication that they represent the risk to human health. It's just that they didn't pass their accreditation process.

So comments, thoughts?

(No response.)

DR. FINDER: Okay. Moving on.

DR. BARR: And the other recommendation is close facilities or facilities with revoked certificates must notify patients and referring physicians. Film retention should apply to facilities that close.

IOM says the complaints from patients who were not informed when their facility closed and were unable or unsure how to access mammography records. If facilities are incapable of notification FDA should notify patients and physicians.

Again, I totally sympathize with patients in this situation. We take an active role with the accreditation bodies in helping patients in this situation, but again, I don't know what authority we can have over a closed facility who's out of business. I don't know what exactly we do to make them do these things.

Charlie.

DR. FINDER: Yeah. Dr. Finder speaking.

This is an issue that come sup not that infrequently, and it can be a big impact on patients, but then the question is what can we do under certain circumstances.

Let me backstep. We have guidance out there to inform all facilities that when they are planning to close what steps they should take, and one of the major steps that they should take is to make arrangements for the retention of records and mechanisms so that patients can access those records, and most facilities do that.

We also ask that the facility notify their accreditation body of those steps and also the FDA so in case anybody asks either the accreditation body or contacts us through our hot line, we can tell the patient what steps they need to go through to find their films.

And those systems do work when we have cooperative facilities. However, when we're dealing with a facility that has gone bankrupt, it is very difficult to deal with those situations. Sometimes there's nobody we can talk to, nobody to reach. Sometimes the records are now part of the bankruptcy hearings and are outside of our jurisdiction. If we can find a sympathetic judge and explain the situation, they have in the past made some type of arrangements.

But, again, it's outside of MQSA. It's outside of the facility's hands because they no longer control the films. They're now in the hands of the bankruptcy court.

In other situations, we have the case where the facility has disappeared. We have no idea of what's going on. The films are gone. The facility itself no longer exists, and while we would like to notify patients of this, we have no information to give them in that situation other than to tell them that their films are gone.

So this is a very tough problem when we're dealing with a truly closed facility, and we would certainly like to hear from the committee about any suggestions they may have about what actions we can take under these conditions, and if there's any way to help the situation.

But there's no question that when a facility goes out of business and doesn't take care of the records and goes into bankruptcy or just closes its doors, shuts the doors, locks them, and disappears, that there are problems for patients. The question is what we can do at that point to help them.

DR. MARTIN: Melissa Martin.

Just a question. Is there any way that you track the physicians, the radiologists tied to those facilities that have closed? So that, in other words, he can't close the facility on one corner and go down the street and open the facility on the next corner and start out as a new entity?

DR. FINDER: Right. Usually in the situation where you've got a physician or facility that has multiple locations, what they'll do in those cases is they'll just transfer the films to the other locations. Those are usually not the problems that we have.

It's individual facilities that go out of business or we have had facilities that have multiple sites where the entire organization went out of the business all at once and affected 100,000 patients.

The interesting thing is that most of those are not mammo patients. These are usually large radiology practices or medical specialty practices. It's not just the mammography records.

We have in those cases worked with the state because they have their own state laws that they can enforce sometimes, and they find it very difficult, too because if they're gone, they're gone. It's very hard to track some of these people.

DR. BARR: I know some states have a requirement that facilities put a bond in case this happens and they can use the money, and I'm not sure at the federal level if we can do that.

I'm wondering if this is something just to make it work really needs to be a state issue.

DR. FINDER: Again, if anybody has any other suggestions, we'd be more than happy to hear about it because, as I say, it's infrequent. When it does occur, it's not pleasant for anybody involved.

DR. FERGUSON: You said there was guidance out there. What is the recommendation for a facility that is closing?

DR. FINDER: The recommendations that we have basically are that the facility inform the state, inform our facility hot line that they're closing; that they make arrangements for those films to be available. And what we suggest is the first suggestion is that they be transferred to another actively and active mammography or radiology facility so that patients can go there and get those films.

If that's not available, we suggest that they go and put them into some type of storage facility where the patients can have access, and as part of that, it's very important that they have some mechanism to inform their patients of what's going on so that it's not just that they do this and nobody knows about it so that patients can't get it.

So we suggest that they have either something on their phone line, an answering machine that gives this message out or that they send out some type of notification or at a minimum that they notify their accreditation body and us so in case patients call us we will know how to forward that information along.

So those are the basic recommendations to the facilities that close. The problem is that if they don't, there's not much we can do after the fact.

MS. RINELLA: Question. Diane Rinella.

What percentage of time do you find that this happens where they actually truly notify that they are going to close versus just closing the doors and going?

DR. FINDER: I can't give you actual numbers. I do know that we get in the mail from facilities this type of information. I can't give you the exact numbers. I will tell you that the number of facilities that close and leave problems like this is a relatively small handful, but when it does occur even at a small size facility, you're probably talking about thousands of patients being affected. So in that sense it is a big issue, and we have had some major facilities that have closed where I think at most there was a group out in California with millions of records, but most of them were not mammography. They were everybody's exams, CTs, medical records of all kinds.

So it's more than jut mammography, and we found that that's usually the case. It's not just a single mammography facility that closes down. It's either a radiology practice so that you've got other patients that aren't even affected or don't have any recourse to us at all for those exams.

MS. PURA: Dr. Barr, that suggestion you had about the bond issue, have states done that before?

DR. BARR: I believe that there are states that have done that. I'm trying to recall. Michigan comes to mind, but I don't want to speak out of turn that they have done it.

We have worked with states and some state have done that. But as I said, at the federal level I'm not sure that that's an option.

Charlie, do you remember?

DR. FINDER: I'm not sure about which, if any, states have instituted a bond. I know that it has been talked about even at this committee, and one of the issues that was brought up is if you're going to have all facilities post the bond when they start, that's another disincentive. It's another issue about, another burden on the facility, considering the fact that the vast majority of facilities that do close do take care of this issue appropriately without anything.

DR. BARR: And I think that's something we have to keep in mind. Since we deal in the public health, we also have to look at is the energy spent on this, although it is very adverse to the patients affected, you know, how do we deal with it in a public health sense and is it a few bad actors and would the state be better able to deal with this and also it's a difficult problem.

MS. PURA: Oh, I know because we go through it quite a bit in Los Angeles and of late it has been a major problem.

DR. FINDER: You know, there are two aspects. One is do you put some type of requirement on all facilities with the idea that in the event that if this happens you can then use that as a fund to accomplish something, although even there you're never 100 percent sure because when they close, we have had the following situations happen where they have their records in some type of filing system that is well known to them. Of course, they're no longer around, and now you can't even categorize these films anymore and hand them out even if you could.

The other is to try and take action against the people that have gone and closed shop, and as I say, when possible we have tried to deal with the bankruptcy court, and in the cases where we have been able to deal with them, they have been receptive and taken some actions.

The states also have power in this area, and recently we did have a state go after a facility and force them to reopen and distribute the films. So it can be done. It's a very tough problem, and I think, as Dr. Barr mentioned, probably our best bet right now is to try and work in conjunction with the state to deal with these problems, but it is a tough one.

DR. BARR: We were just hoping someone would have a brilliant idea we hadn't thought of yet.

And despite Dr. Finder's being sure I wouldn't get done, I finished the section before the break. So, Dr. Hendricks, is it time?

CHAIRPERSON HENDRICKS: I think it is a good time. We'll take a 15 minute break and reconvene at quarter till.

Thank you.

(Whereupon, the foregoing matter went off the record at 3:28 p.m. and went back on the record at 3:50 p.m.)

CHAIRPERSON HENDRICKS: We'll reconvene for the final session this afternoon, which is Dr. Barr on a marathon leading us through the two final topics on work force and beyond mammography.

DR. BARR: Thank you.

I will try to get through this last section. I need to leave here about 4:30-ish. So if we're not finished, then either Mr. Divine will come up here and continue or Dr. Finder can switch seats, but we'll see what we can do.

The next major category for recommendations from the IOM report falls under adequate work force for screening and diagnosis. Under here are Recommendations 7, 8, and 9.

Recommendation 7 is to collect and analyze data on the mammography work force and service capacity; eight, device strategies to recruit and retain highly skilled breast imaging professionals; and, nine, make more effective use of breast imaging specialists.

And we'll start with data on the mammography work force and service capacity. IOM recommends that volume information be collected during annual inspection. HRSA reports on mammography volume by region, state, and type of service. And I think they mean that their reports, which we could contribute to, should include number of facilities, number of mammography units per 10,000 women, number of FTE physicians reading mammograms per 10,000 women stratified by type of service where appropriate.

That we provide unique identifiers for all interpreting physicians, technologists, and medical physicists to get volume services by individual.

That we collect data by facility and waiting times for screening and diagnostic appointments.

That Congress, I assume, not FDA, provide funding to HRSA to model future work force supply and demand on a regular basis.

And the rationale for these recommendations which then we can go back to is to be able to assess accurate real time data to monitor and track capacity on a national and regional basis; to be able to assess the status of the work force; assess appointment waiting times, and assess impact of new regulations and voluntary programs.

So we'll go back to the volume information. Currently the information we have on volume is provided by the facilities to their accrediting bodies on an every three year basis when they apply for reaccreditation, and I don't know how it is in other facilities, but I can certainly tell you when I was practicing, I didn't write down on the form an exact number. I would certain give my best guesstimate of the number of mammograms we performed.

So it has always been debatable how accurate the information we have when we are asked to give volume statistics.

I don't know if that would change if the inspector was asking the question versus the accrediting body, but we'd like to hear your thoughts on how important this information is and the best way to collect it.

Okay. No comments? Okay. Yes?

MS. PURA: Would this invitation by any chance be helpful to such as the CDC, et cetera, to expend monies for reimbursement if we knew that there were so many mammogram units per facility, et cetera, across the country and the ratio of staff to -- do you think that would be helpful, Dr. Barr, or do you think that would be helpful?

I'm always pushing for reimbursement.

DR. BARR: It's really hard to gauge. I myself have been to CMS talking about the costs of at least meeting MQSA regulations if nothing else, and it doesn't seem to affect their reimbursement.

You know, certainly one would think intuitively that more information would inform them to make decisions, but I can't say for sure.

DR. FERGUSON: Does ACR not already kind of have this information about facilities and locations? Would it be easy for them to, I guess, interpolate the number of patients in a geographic area?

DR. BARR: I think the volume information they have is through facilities, but we'll let Penny. I mean through facilities on the accreditation form, but we'll let Penny comment.

MS. BUTLER: Penny Butler from American College of Radiology.

We do collect information, but they're annual patients examined and breakdowns between diagnostics and screens, but we don't have really good, reliable FTE information and some of the other information that is asked for here.

DR. BARR: And certainly asking a volume question during inspection is not a big deal. I mean the inspector could spend two seconds asking that. I think probably the bigger issue here I would say is this unique identifier information. I mean I think that's really where it's at if we want to get some of the data that's being recommended here.

Some of the data that's being recommended here, I can say that from our standpoint there is the problem that if we go to a unique identifier system, then our database has to meet the Privacy Act requirements which it does not have to because we collect data based on facilities. We search by facilities.

And so there would be work and possibly financial burden associated with that. The other thing is, you know, the whole discoverability stuff. Could this be tied into how much mammography you do and what your results are, et cetera?

So just be anxious to hear your comments on B, the unique identifiers, so that we can get volume by individual and perhaps other data.

MR. PASSETTI: Bill Passetti.

I just think it would be nice if some of this standard information was collected on a nationwide basis. In Florida we're having a situation where our legislature is looking at mammography accessibility and different things, and they're already talking about requiring us to collect volume data or how many exams are performed per machine and all this type of data that would be nice to have, but it would be nicer to have on a nationwide basis and not just our state.

DR. BARR: Right. Understandable.

DR. MARTIN: I would just like to -- Melissa Martin -- to take an idea. If you're developing a database to collect this data, put the thought in that if you're going to assign all of the providers' unique identifier numbers, it would also be the option of providing at least on an optional basis the qualifications so that we don't have to kill the trees and provide that.

If we're going to be qualified, then give us one qualification, that we've got all of our CEUs and continuing experience every two years and update it so that we're not having to copy all of the paper work for every facility. Either our physician is going to be qualified or our physicist is going to be qualified or our technologist is going to be qualified.

We're qualified once. It's not going to matter where we're qualified in 15 different facilities.

DR. BARR: Thank you.

DR. FERGUSON: I agree. That was brought up at our last meeting as well. There ought to be a way where we don't have to produce every piece of paper at every facility and have a book that thick. It would be helpful.

CHAIRPERSON HENDRICKS: Comment from the audience?

MS. WILCOX: Pam Wilcox, ACR.

I think there's more complexity to this than even what we're talking about here because even if you look at interpreting physicians and you look at their screening and diagnostic volumes, that doesn't address the capacity of the system because you still have those who are doing biopsies and those who are not doing biopsies. You have patient mix.

There's a whole lot of variables that would make this -- although really it would be great to know and be able to predict where we need to be and what we need to recruit, I'm not so sure it's as simple as it seems even here.

DR. BARR: Thank you.

MR. MOURAD: Wally Mourad, FDA.

During inspections we do download during the previous inspection where a certain person has met the initial qualifications or not, and so those are there, and we don't recheck them every time except for expiring items, like license or approval letter or something like that.

So the only thing we check on continuously every time that's different is the continuing requirements, education and experience. We don't have a database for that. That's the other thing.

DR. FINDER: Right. Dr. Finder.

I just wanted to also mention that for medical physicists because we consider them a special group of -- no, we don't.

(Laughter.)

DR. FINDER: Because they do go to so many different facilities usually, much more than the other personnel categories, and because their requirements tended to be more complex in terms of the documentation that they need, we actually will supply medical physicists a letter stating that they meet all of the initial qualifications, and all they have to do is show that letter to any of their facilities. Give them a copy of that letter, and that is acceptable for all their initial qualifications.

So we have attempted to address some of those issues through that mechanism. We have looked at the issue about having a database with individuals. It does raise a number of issues. One is how do you identify them.

You'd have to assign numbers. It would become a privacy system, but even more so than that there is the issue about how do you get the data in; who do you give that data to; and even if you were allowed and we could figure out a mechanism to give it to our inspectors, what would be the mechanism to give it out to the facilities because it wouldn't do much good if our inspectors knew that you were qualified, but when you showed up at the facility you had no documentation and they would have no way of verifying that.

That's one of the situations we've got right now with the hurricane where people have no documentation. They're showing up at new facilities, and we in this case are providing them with some documentation because of the problems that they have encountered, but we don't give it to the facility.

And then the other issue is how would we deal with the situation where somebody doesn't give us the information at some point in time. What would we do then? Would we search them out and cite them in all of the facilities that they're out because they haven't provided us with data?

So the reason that it keeps coming up is because it would certainly be more convenient if we had this type of system, but there are always these problems that come up that seem to make it difficult to actually implement. And I guess at one point we can take a look at this, have a meeting, and discuss this in detail to see if we can actually implement something like this, but it's not very simple to do.

MR. FLATER: Flater with Iowa.

Just a couple of points. Number one --

CHAIRPERSON HENDRICKS: I want to remind all of the speakers at the microphone to please identify yourself for the purposes of the transcript being prepared of this meeting.

MR. FLATER: Flater, with Iowa.

Just a little bit of a point. Us small people that do everything, including accreditation and certification do have all of the information you're talking about on every place in the State of Iowa, and some physicists outside the state that come in and do some work within our state.

Another thing you might want to look at is the new thing being set up by the Nuclear Regulatory Commission where they're tracking all kinds of sources and everything. There are systems available that will track everything everybody does.

DR. BARR: Thank you.

Recommendation 8 is for strategies to recruit and retain highly skilled professionals. First is to encourage federal and state agencies and health care payers to develop incentives to recruit and retain skilled breast imagers.

Loan repayment awards through the National Health Service Corps, and for J-1 visa waivers for physicians working in underserved areas.

And, again, HRSA should identify and designate shortage areas for breast imaging.

IOM says that existing supplied physicians who read mammograms are high level performance, is a valuable resource, is unproductive to invest in efforts to increase the number of entrants without addressing factors that lead to early departures.

Retaining highly skilled practitioners should be cost effective way to maintain high quality breast imaging services, and the NHSC program that J-1 waivers have been used to bolster work force in other shortage areas.

I think we've heard some of this throughout the discussion today. I think when Dr. Lee from NCR spoke, she addressed some of these issues, as did Dr. Bassett.

Are there any new thoughts about what incentives would work to recruit and retain qualified personnel? Anything that we may be able to do to stop this steady bleed of people leaving the field or not going into the field?

Okay. No new ideas. Again, a tough issue, but certainly what we've heard is that more burden would not help the situation.

IOM says we need to encourage federal and state agencies and health care payers to develop incentives. I think I already did that. Sorry.

IOM says we should support the radiologist assistant training programs and new roles for radiology assistants in breast imaging; that this career option for skilled technologists is an incentive for new entrants and could improve quality, productivity, and efficiency.

And I know we heard at least one, if not more, comments in the public speakers on this. I'd be interested to know if the committee has any thoughts on the radiology assistant area, particularly related to mammography.

Anybody here have experience? Anybody using this type of --

MS. RINELLA: Diane Rinella.

I don't know if any of the radiology assistant programs to date have a specialty for breast imaging, and something tells me they don't, and for right now, the radiology assistants are going through a very comprehensive program for two years for fluoroscopy. You know, so barium enemas and whatnot, and somebody that wants to focus on breast imaging, I don't think that that's something that they would need to or even want to go through.

So I would think that there would need to be something if the RAs were going to be used in the future. A training program specific, an RA format for breast imaging technologists.

DR. BARR: As a technologist and someone experienced in this field and talks to a lot of technologists, do you think that something like this program to go into would make the field more enticing?

MS. RINELLA: It really is very new still. A lot of techs out there still don't know what an RA is, but I have been to I would say a handful of facilities, and the text word going to -- that was one of their goals, go through the RA program -- but it was to become a full blown radiologist assistant. It really wasn't something that was dedicated to mammography.

So I can't speak on just specifically breast imaging.

MS. MOUNT: Carol Mount.

I don't specifically at our facility. There are about three that are just finishing up their bachelor program, and breast imaging is what they would want to do. They've had a lot of questions for me. Of course, I do not have the answers as to what would they be able to do once they were in that field.

And I agree with Diane. There would need to be a specific modality training course in order for it to be effective.

DR. BARR: Thank you.

CHAIRPERSON HENDRICKS: Dr. Monticciolo.

DR. MONTICCIOLO: Debbie Monticciolo.

I guess I have a little bit of a unique perspective because I have worked with a radiologist assistant in a private practice in one of my past jobs, but I would say that I think unless the technologists are willing to accept the medical legal burden, that many radiologists would be, I think, hesitant to pay another individual to help them, but not take the legal burden because that's a huge issue for radiologists.

I don't know if I should say this or not, but I guess I'll just go forward. I worked with a radiologist assistant, and he was very good. I thought he was excellent and was an asset to our practice, but I believe that he was over-utilized by some of the radiologists, and that the radiologists didn't oversee his work as closely as you would expect.

And so that also is another side of this issue that would probably need to be looked at.

DR. BARR: Thank you.

Can ACR help us out on this?

MS. WILCOX: Kim Wilcox, ACR.

There has been an agreement to the responsibilities of a radiologist assistant, and that has been agreed to by the American Registry of Radiologic Technologists, which will be the certifying agency, the American Society of Radiologic Technologists and the ACR, and there will be no interpreting on the part of the radiologist assistant.

So while they might be able to help in the biopsy area or some of these other areas, I'm not sure what they would do to help the radiologist shortage that we have in breast imaging right now.

And as was said this morning and as Debbie reiterated, the medical liability issue is huge.

DR. BARR: Thank you. That really helps a lot.

DR. FLATER: Flater with Iowa.

Number one, if we did anything with the radiologist assistant, we would have to completely change our rules because interpreting physicians in Iowa must be radiologists. There's no choice.

The other side of it that's coming up that we're going to talk to later is in the stereotactic. Dr. Finder has just spent some time with us in Iowa because we have an RA in training right now that is trying to get us to agree to allow him to go through the training program to do stereotactic. There are radiologists that are willing to do training.

So that's an issue we're trying to deal with right now. Today we wouldn't allow it.

DR. BARR: Thank you.

MS. PURA: Dr. Barr.

DR. BARR: Yes.

MS. PURA: Linda Pura.

DR. BARR: Oh, I'm sorry, Linda.

MS. PURA: You know, these are the growing pains that physician assistants and nurse practitioners had in the past, and we all know that, and certainly this might be something to look at to give futuristic kind of career growth for the rad techs, and I don't know if they're not allowed to interpret at this point now, would they be able to do secondary reading for the double reading? Would that be a possibility?

DR. BARR: I think those are all good points, and it sounds like in this area we probably don't have enough information yet of how this is going to and how these people can possibly help us in the mammography area.

So I think as this goes on we'll certainly have to keep an eye on this program and see where it can possibly be of use to radiologists and as an incentive.

Good points. I think we have models probably of other professions having gone through this. It's a good point.

IOM says that support should be given to demonstration projects to evaluate potential for double reading by non-physician clinicians, and again, this is based on the rationale that double reading has the potential to improve interpretation and perhaps that's an area where the RA would fit in.

And to evaluate the roles of ancillary personnel and mammography, productivity will be maximized according to IOM if radiologic technologists focus on performing mammograms and interpreting physicians' focus on interpretation, ancillary personnel, technical and nontechnical responsibilities, including quality control and administration.

Now, when I first read this, I was like, "Well, gee whiz. You know, we have this whole QC tech thing where you have to be in this area," but I'm not sure that's exactly what they mean.

Charlie, I know when you read this, you had a little different thought about what IOM might be saying here.

DR. FINDER: Well, I think what they're trying to get at is the idea of basically having the radiologic technologists spend all of their time doing patients, interpreting physicians just doing the interpretations and leave any of the other paper work, quality control areas to nontechnologists, nonprofessionals or nonpersonnel, as we define personnel.

DR. BARR: With oversight.

DR. FINDER: Including quality control because we do allow personnel or people other than RTs to do the QC procedures. As long as they're under the supervision of a quality control technologist, other people who have received adequate training can perform these various tests.

So I think what they're trying to get at is they realize that there are shortages for the techs and for the physicians, and to focus them on just doing the aspects of mammography that only they can do.

DR. BARR: And as you say, MQSA does allow for this oversight type of responsibility at the current time.

Next is this new topic is improving breast imaging quality beyond mammography. Any comments on any of the personnel incentives before we move on to these final comments?

(No response.)

DR. BARR: Okay. Recommendation 10 in the report is accreditation for non-mammography breast imaging modalities, such as ultrasound and MRI. The rationale is accreditation already exists for breast ultrasound and general MRI and a breast specific MRI accreditation program is under discussion.

Accreditation for breast imaging methods would lead to standardization and improved quality of breast cancer detection and diagnosis.

I think we heard earlier in some of the talks that the MRI accreditation or anything in federal regulation of breast MRI probably isn't in the immediate future, but that perhaps breast ultrasound is an area for exploration.

Any comments from the committee?

MS. RINELLA: Diane Rinella.

I am also RDMS in breast ultrasound and in my travels I do question the facilities and who is doing their breast ultrasound examinations, and I asked what type of equipment, what types of transducers and things that they're using, and it's unfortunate that they aren't all up to the same standard of care.

And so I would support standardization of and accreditation for breast ultrasound just because of what I'm seeing out there in the field.

DR. BARR: Thank you.

Anybody else?

(No response.)

DR. BARR: I guess that's it. Yeah, I think that's it, except for the stereotactic section, which we'll discuss tomorrow. The recommendations, I think we've pretty much gone through the recommendations.

One thing I wanted to point out is that the final printed version of the IOM report actually just came out. So although there are no major changes, there could be like when Penny Butler was up here talking about the BI-RADS thing. There could be wording changes that in the draft were one way so that we have them on our slides, but in the final report have come out slightly differently.

Charlie.

DR. FINDER: No, I just want to say the version that you have is the current one now.

CHAIRPERSON HENDRICKS: That's going to bring to a conclusion a long day.

I do have one housekeeping detail. The woman who is transcribing the meeting has had difficulty throughout the day today understanding the names of the speakers who have come from the audience. So she requested of those speakers to get full recognition of your comments just stop by her desk to clarify the spelling of your first and last name.

Thank you.

Otherwise, this concludes the meeting, and we will reconvene again tomorrow morning at 9:00 a.m.

(Whereupon, at 4:18 p.m., the meeting was adjourned, to reconvene at 9:00 a.m., Tuesday, September 27, 2005.)